Multi-angle defect detection method for batch chips, control equipment and AOI (Automatic Optic Inspection) equipment

Abstract

The invention relates to an optical measurement technology, and discloses a batch chip multi-angle defect detection method, control equipment and AOI equipment, and the method comprises the steps: placing a chip array after gold wire bonding into a detection area of the AOI equipment; controlling the camera array to shoot panoramic images of the chip array in multiple directions; carrying out image segmentation on the panoramic image in multiple directions to obtain a multi-angle image corresponding to each chip; carrying out defect detection on the chip without the unqualified image; an image supplementary shooting path is intelligently planned according to the chip position and direction to which the unqualified image belongs; controlling the motion camera to carry out supplementary shooting on the unqualified image along the image supplementary shooting path; carrying out defect detection on the chip after image supplementary shooting; and generating a defect detection result of the chip. The invention further discloses a computer readable storage medium. The invention aims to improve the efficiency of defect detection on batch chips while ensuring the quality of defect detection on the batch chips.

Description

Technical Field

[0001] The present application relates to the field of optical measurement technology, and in particular to a multi-angle defect detection method for batch chips, a control device, an AOI device, and a computer-readable storage medium. Background Art

[0002] In the modern semiconductor manufacturing industry, chip quality inspection is a key link to ensure product reliability and performance. With the continuous improvement of chip integration and complexity, the efficiency and accuracy requirements for chip defect detection are also getting higher and higher. As an efficient inspection method, Automated Optical Inspection (AOI) is widely used in the chip manufacturing process to ensure chip quality and production efficiency.

[0003] At present, the way to use AOI equipment to automatically analyze chips for defects is generally to take corresponding chip images one by one in each direction. Each chip and each direction needs to be photographed independently. This process requires the camera to constantly turn and stop between different directions to ensure comprehensive and detailed image acquisition of each direction of each chip.

[0004] Although this method of shooting chip by chip and position by position ensures the comprehensiveness and accuracy of the image to a certain extent, it also brings some problems: First, since each chip and each position needs to be photographed independently, the overall inspection time is too long. Especially in the case of large-scale chip inspection, this method of shooting one by one greatly prolongs the inspection cycle and affects production efficiency; secondly, during the shooting process, the frequent turning and stopping of the camera may cause unstable movement, which in turn affects the clarity and consistency of the image. In addition, the frequent turning and stopping also increase the mechanical wear of the camera clamping parts, reducing the stability and service life of the equipment.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a multi-angle defect detection method for batch chips, a control device, an AOI device and a computer-readable storage medium, aiming to ensure the quality of defect detection for batch chips while improving the efficiency of defect detection for batch chips.

[0007] To achieve the above objectives, the present application provides a multi-angle defect detection method for batch chips, comprising the following steps: The chip array after gold wire bonding is placed in the detection area of ​​the AOI equipment; wherein the detection area is provided with a camera array and a motion camera; Control the camera array to operate at multiple shooting angles to capture panoramic images of the chip array in multiple directions; during the camera array shooting process, each camera focuses independently; Perform image segmentation on panoramic images in multiple directions to obtain multi-angle images corresponding to each chip; Perform image quality checks on images from multiple angles and filter out unqualified images; For chips without unqualified images, multi-angle images are stitched to obtain the overall image of the chip, and the overall image of the chip is analyzed based on the AOI technology to detect defects on the chip; and the image reshoot path is intelligently planned according to the chip position and orientation to which the unqualified image belongs; the path in the same orientation is calculated according to the shortest distance of the connecting line of the chip positions to which all unqualified images in the same orientation belong; and the paths in different orientations are connected to form the image reshoot path according to the shortest switching distance between the paths in different orientations; Control the motion camera to retake the unqualified images along the image retake path; wherein the interval path of the motion camera between the two retake points includes a speed-up segment, a speed-down segment, a uniform speed preparation segment and a uniform speed capture segment in sequence, and the length of the uniform speed preparation segment is positively correlated with the deceleration of the speed-down segment; the end point of the uniform speed capture segment of the previous interval path is the start point of the speed-up segment of the next interval path; Replace unqualified images with retaken images; For the chip after the image reshoot, multi-angle images are stitched together to obtain the overall image of the chip, and the overall image of the chip is analyzed based on AOI technology to detect defects on the chip; Mark non-defective chips as qualified products, mark defective chips as unqualified products, mark chips that fail to complete defect detection as pending confirmation products, and generate defect detection results for the chips.

[0008] To achieve the above-mentioned purpose, the present application also provides a control device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, the steps of the multi-angle defect detection method for batch chips as described above are implemented.

[0009] To achieve the above-mentioned purpose, the present application further provides an AOI device, the AOI device comprising a camera array, a motion camera and a control device; wherein the camera array and the motion camera both establish a communication connection with the control device and are controlled by the control device; The control device is the control device as described above.

[0010] To achieve the above objectives, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the multi-angle defect detection method for batch chips as described above are implemented.

[0011] The multi-angle defect detection method, control device, AOI device and computer-readable storage medium for batch chips provided by the present application place the chip array after gold wire bonding into the detection area of ​​the AOI device, and use the camera array to simultaneously shoot multiple angles of multiple chips, which significantly reduces the independent shooting time of each chip and improves the overall detection efficiency. While performing chip defect detection on qualified images, the motion camera can also be used to intelligently reshoot some images of insufficient quality taken by the camera array to ensure the quality and efficiency of detection. In addition, during the reshooting process, unnecessary movements and pauses are reduced by intelligently planning the path and optimizing the motion control of the camera, which not only takes into account the reshooting efficiency and the quality of the reshooting images, but also reduces mechanical wear and extends the service life of the equipment. In this way, it helps to improve the overall defect detection efficiency of batch chips, thereby improving the production efficiency of batch chips and ensuring chip quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a schematic diagram of the steps of a multi-angle defect detection method for batch chips in one embodiment of the present application; Figure 2 This is a schematic diagram of the speed of each segment of the interval path in an embodiment of the present application; Figure 3 Schematic diagram of the internal structure of a control device according to an embodiment of the present application.

[0013] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0014] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0015] In addition, if the descriptions of "first", "second", etc. are involved in this application, they are only used for descriptive purposes (such as for distinguishing the same or similar features), and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in this field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0016] Reference Figure 1 In one embodiment, a multi-angle defect detection method for batch chips includes: Step S10, placing the chip array after gold wire bonding into the detection area of ​​AOI equipment; wherein the detection area is provided with a camera array and a motion camera; Step S20, controlling the camera array to operate at multiple shooting angles to capture panoramic images of the chip array at multiple positions; wherein, during the camera array shooting process, each camera focuses independently; Step S30, performing image segmentation on the panoramic images in multiple directions to obtain multi-angle images corresponding to each chip; Step S40, performing image quality inspection on the multi-angle images to filter out unqualified images; Step S50: for chips without unqualified images, stitching of multi-angle images is performed to obtain an overall image of the chip, and image analysis of the overall image of the chip is performed based on the AOI technology to detect defects on the chip; and, according to the chip position and orientation to which the unqualified image belongs, an image reshooting path is intelligently planned; wherein, the path in the same orientation is calculated according to the shortest distance of the connecting line of the chip positions to which all unqualified images in the same orientation belong; and the paths in different orientations are connected to form an image reshooting path according to the shortest switching distance between the paths in different orientations; Step S60, controlling the motion camera to retake the unqualified images along the image retake path; wherein the interval path of the motion camera between the two retake points includes a speed-up segment, a speed-down segment, a uniform speed preparation segment and a uniform speed capture segment in sequence, and the length of the uniform speed preparation segment is positively correlated with the deceleration of the speed-down segment; the end point of the uniform speed capture segment of the previous interval path is the start point of the speed-up segment of the next interval path; Step S70, replacing the unqualified image with the retaken image; Step S80: stitching the multi-angle images of the chip after the image reshoot to obtain the overall image of the chip, and performing image analysis on the overall image of the chip based on the AOI technology to perform defect detection on the chip; Step S90: Mark non-defective chips as qualified products, mark defective chips as unqualified products, mark chips that have failed defect detection as pending products, and generate defect detection results for the chips.

[0017] In this embodiment, the execution terminal of the embodiment may be a control device of the AOI device.

[0018] As described in step S10, before entering this step, the chip has been assembled by a gold wire bonding process. Gold wire bonding generally connects thin metal wires to a substrate or carrier to achieve electrical connection. This process is called bonding welding wiring, and usually uses thermal compression bonding or ultrasonic bonding technology to weld metal wires between pads on the chip and pins on the substrate.

[0019] The gold wire bonded chips are placed in an array (usually in a row and column format) on a dedicated tray or carrier. The position of each chip in the array needs to be accurately identified for positioning and detection in subsequent steps.

[0020] Optionally, an operator or an automated system of an AOI device (such as a gold wire bonding AOI sorter) moves the carrier with the chip array smoothly into the inspection area of ​​the AOI device. The carrier can then be transported to the shooting position by a conveyor belt, a robotic arm, or other automated equipment. In the inspection area of ​​the AOI device, a camera array and a motion camera are set.

[0021] Optionally, the camera array can be composed of multiple high-resolution cameras, the specific number of which depends on the detection requirements and chip size (it is not necessary to require that the number of cameras in the camera array is consistent with the number of chips, and the shooting range of one camera can cover multiple cameras). For example, there can be 4, 8 or more cameras to cover different angles of the chip.

[0022] Optionally, the motion camera is mounted on a movable transport platform, which consists of precision servo motors and a high-precision rail system, and can move freely within the inspection area. The motion camera can be a high-resolution single camera or a multi-camera system, and the specific configuration depends on the inspection requirements. The motion camera has enough movement range to cover all areas of the chip array.

[0023] Of course, the camera array also has a certain degree of motion capability (the range of motion and flexibility do not have to reach the level of sports cameras) so that it can operate multiple shooting angles when shooting. Among them, the movement of the camera can be controlled by an electric motor or a stepper motor to achieve precise position control; some cameras can also have a height adjustment function, allowing them to move up and down within a certain range to adapt to chips or detection requirements of different heights.

[0024] As described in step S20, the initial position of the camera array may be located directly above the camera array, and a top-down panoramic image of the camera array is first acquired.

[0025] Optionally, ensure that each camera in the camera array is in an initial position, usually a top-down position. Each camera needs to preset an initial focus point to ensure that it can focus quickly during shooting. Each camera performs a focus operation independently to ensure that the captured image is clear. Then, the images captured by each camera are stitched to obtain a panoramic image of the chip array at a top-down angle.

[0026] Then, according to the required shooting angle, the movement path of the camera array is planned to switch to the next shooting angle, such as a bird's-eye view, a side view, etc. The camera array can be controlled by an electric motor or a stepper motor to adjust the position and angle of the camera array to ensure that each camera can accurately reach the predetermined shooting angle. At each new angle, each camera performs focus operation again independently to ensure a clear image at the new shooting angle.

[0027] Taking bird's-eye view shooting as an example, it can be divided into bird's-eye view angles in front, behind, left and right of the chip. After adjusting the camera array to the bird's-eye view shooting angle of the corresponding orientation, each camera performs focus operation independently to ensure that the captured image is clear. Then the chip array images taken by each camera at the bird's-eye view angle of the corresponding orientation are spliced ​​to obtain a panoramic image at the bird's-eye view angle of the corresponding orientation. These images will be used for subsequent image processing and analysis.

[0028] For example, each camera captures a local image of the chip array at a bird's-eye view in front of the chip array, and then stitches the captured images to obtain a panoramic image at a front bird's-eye view; then, the camera array is controlled by an electric motor or a stepper motor, and after adjusting to a bird's-eye view on the left side of the chip array, each camera captures a local image of the chip array at a bird's-eye view on the left side of the chip array, and then stitches the captured images to obtain a panoramic image at a left bird's-eye view.

[0029] Optionally, if images at side view angles (left, right, front, and back) are required, the position and angle of the camera array can be adjusted. At each new side view angle, each camera performs focus operation again independently to ensure that the image is clear at the new shooting angle, and to capture a panoramic image of the chip array at the side view angle.

[0030] In this way, multi-directional panoramic images of the chip array can be effectively acquired, providing high-quality basic data support for subsequent image processing and defect detection.

[0031] As described in step S30, after the camera array captures a panoramic image of the chip array from multiple angles (a panoramic image is a large image containing multiple chips, each chip occupies a specific position in the image), the panoramic images at different angles can be segmented based on a single chip.

[0032] Optionally, before image segmentation, some preprocessing operations can be performed on the panoramic image to improve the accuracy and efficiency of segmentation. The preprocessing steps can include: removing noise in the image to reduce interference in the segmentation process; enhancing the contrast of the image to make the edge of the chip clearer to facilitate subsequent segmentation operations.

[0033] The purpose of image segmentation is to separate each chip in the panoramic image and obtain an independent image for each chip. Optional image segmentation algorithms include: using the Canny edge detection algorithm or other edge detection methods to identify the edge of the chip; or by setting a suitable threshold to separate the chip area from the background area in the image.

[0034] In the panoramic image, the position of each chip is known (it may have been marked in step S10). Based on this position information, the segmentation operation can be performed more accurately. The coordinate range of each chip in the panoramic image can be determined based on the pre-marked chip position, and then the image of each chip can be extracted from the panoramic image based on the coordinate range of the chip using image processing tools (such as OpenCV).

[0035] Since panoramic images are taken from multiple angles, each chip has a corresponding image at different angles. During the segmentation process, it is necessary to ensure that the image of each chip at each angle is correctly extracted. Therefore, the panoramic image can be classified according to the shooting angle (such as top view, bird's eye view, side view, etc.), and then the panoramic image at each angle is segmented to extract the image of each chip at that angle.

[0036] As described in step S40, although the camera array can be used to shoot multiple angles of multiple chips at one time, the independent shooting time of each chip can be significantly reduced, thereby improving the overall shooting efficiency. However, due to the arrangement of the camera array, for chips at certain positions or angles, it is inevitable that the limited field of view is insufficient, resulting in poor image quality (for example, chips at certain edge positions or specific angles may not be fully covered or blocked, or images shot at certain positions may be affected by optical distortion, resulting in distorted or blurred image edges), so it is necessary to filter out images of poor quality.

[0037] That is, the purpose of image quality inspection is to screen out images that do not meet quality standards due to poor shooting conditions, equipment failure or other reasons. These unqualified images may affect subsequent defect detection results, so they need to be screened before further processing.

[0038] When performing image quality inspection, multiple evaluation indicators can be used to judge the quality of the image. Commonly used evaluation indicators include: (1) Clarity: Image clarity is one of the important indicators for measuring image quality. A clear image can provide more detailed information, which is convenient for subsequent defect detection. (2) Contrast: The contrast of an image reflects the brightness difference between different regions in the image. Images with high contrast make it easier to distinguish different regions and features. (3) Noise level: Noise in an image can interfere with image detail and reduce image quality. (4) Exposure: The exposure of an image reflects the brightness level of the image. Appropriate exposure can ensure that the details in the image are not obscured by overexposure or underexposure. (5) Distortion: Image distortion may be caused by improper camera lens or shooting angle. Distortion will affect the geometric shape of the image, thereby affecting subsequent analysis and detection.

[0039] Among them, the clarity assessment can use edge detection algorithms to calculate the clarity of the image. A clear image usually has a higher edge strength.

[0040] Among them, contrast evaluation can be achieved by calculating the contrast of the image and calculating the grayscale histogram of the image. A high-contrast image will have obvious peaks and valleys in the histogram.

[0041] Among them, noise assessment can evaluate the noise level by calculating the signal-to-noise ratio of the image.

[0042] Among them, exposure assessment can determine whether the image is overexposed or underexposed by calculating the average brightness value of the image.

[0043] Distortion assessment can determine whether an image is distorted by calculating the geometric shape change of the image. Optional methods include Harris corner detection and image registration.

[0044] According to the results of image quality assessment, unqualified images are screened out and marked.

[0045] Optionally, images that do not meet the requirements are screened out according to a preset image quality threshold, for example, images with clarity below a certain threshold, insufficient contrast, excessive noise level, improper exposure or distortion.

[0046] Mark the unqualified images that have been screened out, and record their chip positions and shooting angles for subsequent image reshooting and processing.

[0047] As described in step S50, when the quality of each angle image corresponding to the chip is qualified, it is marked as a chip without unqualified images; when the quality of any angle image of the chip is unqualified, it is marked as a chip with unqualified images.

[0048] Optionally, for chips without unqualified images, feature detection algorithms (such as SIFT, SURF, or ORB) are used to extract feature points from images at each angle and perform matching to find the correspondence between images at each angle.

[0049] Based on the feature matching results, image stitching algorithms (such as stitching, fusion, etc.) are used to merge multi-angle images into a complete chip image. In this process, geometric transformation and color correction can be performed if necessary to ensure the natural and accurate stitching effect.

[0050] Analyze the overall image of the spliced ​​chip and use AOI image processing technology to identify potential defects (machine learning-based visual recognition can be added if necessary). This can include but is not limited to detecting whether there are multiple wires or misaligned wires on the chip leads, and whether the chip body is complete or missing.

[0051] While performing defect detection on chips without unqualified images, the position information and orientation information of all unqualified images are collected for unqualified images (each unqualified image has a clear chip position and shooting angle (orientation)).

[0052] For each specific position, the shortest path from the current position to all unqualified chip positions in that position is calculated. This can be solved using a variant of the Traveling Salesman Problem (TSP) algorithm to ensure that the camera can visit each chip position that needs to be reshot in the shortest time. At the same time, an adjacency matrix is ​​established in which the distance between each pair of chip positions is calculated and stored, so that the motion camera can visit each chip position in turn with the shortest distance.

[0053] In the planned reshoot path, a position corresponding to a chip position is a reshoot point.

[0054] After completing the path planning for each position, you need to consider how to switch from the current position to the next position that needs to be retaken. This involves calculating the shortest switching path from the last chip position of the current position to the first chip position of the next position. Among them, the shortest path from the end point of one position to the starting point of another position can be calculated to ensure that the sports camera can switch positions quickly and efficiently.

[0055] Optionally, the shortest path in the same orientation is connected with the shortest switching paths between different orientations to form a complete path, which ensures that the motion camera can sequentially visit all the positions of the unqualified images that need to be retaken, no matter where they are located.

[0056] In the process of connecting various orientation paths, the total path can be further optimized to ensure that the motion camera completes all reshoot tasks in the shortest possible time with the minimum number of turns.

[0057] As described in step S60, before the reshoot begins, confirm that the action camera is in a ready state, and all necessary parameters (such as exposure time, focal length, etc.) have been pre-set to adapt to the reshoot environment. The previously planned image reshoot path is loaded into the control system of the action camera.

[0058] In the image retake path, the path between two adjacent retake points is defined as an "interval path". Figure 2 ,Each interval path includes a speed-up section, a speed-down section, a uniform speed preparation section and a uniform speed capture section in sequence.

[0059] Optionally, in the acceleration phase, the motion camera starts from a stationary or low speed (when retaking the first image in each orientation path, the initial speed is 0; when retaking the second image, the initial speed is the speed at the end of the uniform speed capture phase), and gradually accelerates to the preset target speed v1. This acceleration process is smooth to reduce mechanical shock and image distortion. The value of the acceleration is determined by the mechanical characteristics of the control platform that controls the motion of the motion camera, and the maximum value can be taken within the range allowed by the mechanical characteristics.

[0060] The deceleration stage is the stage where the motion camera starts to decelerate when approaching the next chip position to ensure that the speed is stable and controllable when reaching the reshoot point. The deceleration value is adjusted as needed to ensure that the camera decelerates smoothly within the predetermined time. Among them, although the deceleration value can also be determined by the mechanical characteristics of the operation and control platform that controls the movement of the motion camera, the maximum value can be taken within the range allowed by the mechanical characteristics, but since the acceleration stage does not need to consider the motion shake too much, and the deceleration stage needs to avoid severe shaking of the operation and control platform or the camera, it is preferred to set the deceleration value to be smaller than the acceleration of the acceleration stage.

[0061] The uniform speed preparation stage is used to alleviate or eliminate the camera shake that may be caused by the deceleration stage. When the action camera decelerates in the deceleration stage, the camera may have a certain shake or vibration due to the effect of mechanical inertia. This shake may affect the clarity and quality of the image; and during the deceleration process, the speed fluctuation may also cause the camera to be unable to stabilize immediately when approaching the retake point, thus affecting the snapshot effect.

[0062] The uniform speed preparation stage provides a stable environment for the camera by maintaining a constant speed, so that it can eliminate the shaking and vibration caused by the deceleration stage before taking a snapshot. The length of the uniform speed preparation stage is positively correlated with the deceleration of the deceleration stage, which means that if the deceleration of the deceleration stage is large, the length of the uniform speed preparation stage will also increase accordingly. This design provides the camera with more time to stabilize the speed, thereby reducing shaking.

[0063] Optionally, the length of the uniform speed preparation section is determined according to the deceleration of the deceleration section. The following formula can be used for calculation: L= ; Among them, v1 is the target speed before deceleration, v2 is the uniform speed after deceleration, a is the deceleration, and k is the empirical coefficient. The value of the empirical coefficient k is usually determined according to the actual application scenario and test data, such as being set between 2 and 10.

[0064] In the uniform speed preparation section, the speed of the motion camera is precisely controlled at a constant value. The control system will calculate the appropriate uniform speed preparation section length based on the deceleration of the deceleration section and the mechanical characteristics of the camera to ensure that the camera reaches a stable state before capturing the image. Even though the distances between the reshoot points may be different, it is still necessary to ensure that the length of the uniform speed capture section of each interval path can ensure that the motion camera has enough time to capture the chip image. Therefore, the length of the uniform speed capture section must be determined based on the actual uniform speed of the uniform speed capture section and the camera parameters of the motion camera.

[0065] Optionally, during the uniform speed preparation phase, the system can also fine-tune the position of the camera to ensure that the camera is positioned accurately when capturing images.

[0066] The uniform speed preparation section design allows the camera to reach a stable state before capturing, significantly improving the clarity and quality of the image. The uniform speed preparation section design not only improves the image quality, but also ensures that the camera can efficiently move and capture between multiple retake points, improving overall operational efficiency.

[0067] Optionally, in actual operation, the length of the uniform speed preparation segment and the speed control strategy may be dynamically adjusted according to the real-time monitored camera status and environmental conditions to meet different reshoot requirements.

[0068] Optionally, the acceleration sensor installed on the action camera is used to monitor the acceleration and vibration of the camera in real time. According to the acceleration data monitored in real time, the acceleration and deceleration of the camera are dynamically adjusted. For example, by optimizing the acceleration curve (such as the S-shaped curve), the acceleration change is smoother and the mechanical shock and shaking are reduced; if it is detected that the camera shakes greatly in the deceleration section, the deceleration can be appropriately reduced or the length of the uniform speed preparation section can be increased.

[0069] The uniform speed capture stage is to capture images when the speed of the motion camera is stable and the position is accurate. In this stage, it is necessary to ensure that the camera can capture the image of the chip quickly and accurately.

[0070] Once the capture of the current chip is completed, the end point of the uniform capture segment immediately becomes the starting point of the next interval path speed-up segment. This design ensures that the camera can move and retake between multiple retake points continuously and efficiently.

[0071] The above process will be repeated on each interval path until all reshoot points are visited and images are successfully reshot. During the entire reshoot process, the system will monitor the status of the camera, the accuracy of the path, and the quality of the reshot images in real time. If any problems or deviations are found, the system will make timely adjustments to ensure the smooth completion of the reshoot task.

[0072] It should be noted that the speed-up section is designed to enable the sports camera to approach the next reshoot point as quickly as possible; the speed-down section is designed to ensure that the speed of the sports camera is stable and controllable when it reaches the reshoot point (that is, to ensure that the camera has a moderate speed when capturing, to avoid image blur due to excessive speed); the uniform speed preparation section is designed to enable the sports camera to enter an absolutely stable and uniform translation state before the formal capture (that is, to alleviate or eliminate the camera shake that may be caused by the speed-down section, and to allow the camera to reach a stable state before capturing by maintaining a constant speed); the uniform speed capture section is designed to enable the sports camera to capture clear, high-quality chip images during the process of stable and uniform translation through the reshoot point.

[0073] In this way, there is no need to pause the motion camera during the capture process, which can not only improve the efficiency of reshooting, but also ensure high-quality chip images.

[0074] The appearance of unqualified quality images is mostly caused by the field of view of the camera array. The appearance of this reason is affected by factors such as the position of the camera array relative to the chip array, the viewing angle, the ambient light, and the focus parameters, resulting in uneven distribution of unqualified quality images, which also makes the positions of the reshoot points unevenly distributed. By designing the interval path between the two reshoot points to include a speed-up section, a speed-down section, a uniform speed preparation section, and a uniform speed capture section in sequence, the motion camera can efficiently move and capture between the unevenly distributed reshoot points while ensuring the image quality.

[0075] As described in step S70, the motion camera retakes the designated chip position and orientation according to the planned path to generate new image data.

[0076] According to the metadata of the retaken image or through image processing technology, the chip position and shooting orientation corresponding to the retaken image are determined. The retaken image is registered with the original image set to ensure that the new image can accurately replace the original unqualified image.

[0077] Optionally, in the image database, find the unqualified image record corresponding to the chip position and orientation, and replace it with the retaken image. Update the metadata information of the image, including shooting time, camera parameters, location information, etc., to reflect the replacement operation.

[0078] As described in step S80, for the chip after the image reshoot, a feature detection algorithm is used to extract feature points from the images at each angle, and matching is performed to find the corresponding relationship between the images at each angle.

[0079] Based on the feature matching results, an image stitching algorithm is used to merge the multi-angle images into a complete chip image. In this process, geometric transformation and color correction can be performed if necessary to ensure the natural and accurate stitching effect.

[0080] Analyze the overall image of the spliced ​​chip and use AOI image processing technology to identify potential defects (machine learning-based visual recognition can be added if necessary). This can include but is not limited to detecting whether there are multiple wires or misaligned wires on the chip leads, and whether the chip body is complete or missing.

[0081] As described in step S90, for those chips that have not been found to have any defects after AOI technology inspection, they are marked as "qualified products". Qualified products mean that the chips have passed all quality inspection standards and can enter the next production link or be packaged and shipped.

[0082] For chips that are detected to have defects, they are marked as "defective" according to the type and severity of the defects. These chips require additional processing, such as repair, scrapping or re-testing.

[0083] For those chips that fail to complete defect detection due to some reasons (such as insufficient light, blurred image, failure to meet the identifiable standards of the defect detection algorithm, etc.), they are marked as "pending confirmation". These chips need to be re-tested or manually confirmed to determine their final quality status.

[0084] Optionally, update the test results of each chip to the database, recording chip number, test time, test results (qualified, unqualified, pending confirmation), defect type, etc. Ensure the traceability and integrity of all data.

[0085] Optionally, the test results are associated with the corresponding chip images to form a complete test report. The test report can include the original image, the stitched panoramic image, the defect mark image, etc., providing detailed quality information.

[0086] Optionally, a detailed quality report can be generated. The report format may include basic information of the chip (such as batch number, production date, etc.), statistical information of the test results (number of qualified products, number of unqualified products, number of products to be confirmed), defect type and distribution chart, defect severity analysis, etc.

[0087] Optionally, use a combination of charts, images and text to generate an intuitive and easy-to-understand quality report. The report should include defect distribution diagrams, defect type ratio diagrams, defect severity analysis diagrams, etc., to help managers quickly understand the quality status of the chip.

[0088] Optionally, the generated quality report is fed back to the production line or quality control department to guide subsequent production adjustments and quality improvements. For unqualified chips, it is recommended to repair or scrap them; for chips to be confirmed, it is recommended to re-test or manually confirm them.

[0089] Optionally, data analysis can be performed on the defect detection results to identify the distribution patterns and possible causes of defects. Through data analysis, potential problems in the production process can be discovered, providing a basis for process improvement and equipment calibration.

[0090] Optionally, archive all generated quality reports and related image files for future review and auditing. Ensure the traceability and integrity of all data to meet the requirements of the quality management system.

[0091] In one embodiment, the chip array after gold wire bonding is placed in the detection area of ​​the AOI equipment, and the camera array is used to simultaneously shoot multiple angles of multiple chips, which significantly reduces the independent shooting time of each chip and improves the overall detection efficiency. While performing chip defect detection on qualified images, the motion camera can also be used to intelligently reshoot some images of insufficient quality taken by the camera array to ensure the quality and efficiency of detection. In addition, during the reshooting process, unnecessary movements and pauses are reduced by intelligently planning the path and optimizing the motion control of the camera, which not only takes into account the reshooting efficiency and the quality of the reshooting images, but also reduces mechanical wear and extends the service life of the equipment. In this way, it helps to improve the overall defect detection efficiency of batch chips, thereby improving the production efficiency of batch chips and ensuring chip quality.

[0092] In one embodiment, based on the above embodiment, the step of performing image analysis on the overall image of the chip based on the AOI technology to perform defect detection on the chip includes: Locate the gold wire image in the overall chip image based on the edge detection algorithm and analyze whether the wiring in the gold wire image is correct; Based on image segmentation technology, the chip body image is separated from the overall chip image, and the chip body image is detected to see if there are defects.

[0093] In this embodiment, when performing defect detection on a chip that does not have an unqualified image or on a chip after image reshooting, image analysis of the entire chip image can be performed based on AOI technology to perform defect detection on the chip.

[0094] Optionally, if necessary, the overall chip image may be subjected to image preprocessing such as denoising, grayscale conversion, and binarization.

[0095] Defect detection includes but is not limited to detecting whether there are multiple wires or misaligned wires on the chip leads, and whether the chip body is complete or missing.

[0096] Optionally, when detecting whether there are multiple wires or misaligned wires in the chip leads, the Canny algorithm can be used to detect the edges in the image, identify the outline of the gold wire, and then detect the straight line through Hough transform to further confirm the position and direction of the gold wire. The outline of the gold wire is extracted from the edge image, and the geometric features of the gold wire, such as length, angle, and position, are analyzed to determine the specific position of the gold wire.

[0097] The predefined standard gold wire template is used to compare with the detected gold wire image, and the difference (such as Euclidean distance, correlation coefficient) is calculated to determine whether the gold wire is correctly connected. The gold wire image is geometrically transformed to align it with the standard template, and then check whether there are extra gold wires or whether the gold wire is connected to the wrong pin or position.

[0098] Optionally, when checking whether the chip body is complete or missing, the chip body image can be separated from the overall chip image based on image segmentation technology. Starting from the seed point, the chip body area can be gradually grown; then the chip body can be separated from the background by setting a suitable threshold; finally, operations such as expansion and erosion can be used to further optimize the segmentation result.

[0099] Optionally, the area and perimeter of the chip body are calculated and compared with standard values ​​to detect whether there is any missing or damaged part. Alternatively, the roundness and shape of the chip body are analyzed to detect whether there is any abnormality.

[0100] Optionally, based on the gray level co-occurrence matrix, the texture features of the chip body are analyzed to detect whether there are abnormal areas.

[0101] Optionally, check whether the chip body is a single connected area and detect whether there are broken or separated parts.

[0102] If the gold wires of the chip are connected correctly and the chip itself is intact and without defects, it is a qualified product; if there are problems such as multiple lines, wrong lines, and / or the chip itself is missing, damaged, or has abnormal textures, it is an unqualified product; if it is impossible to determine whether it is qualified due to image quality problems or other reasons and manual re-inspection is required, it is a product to be confirmed.

[0103] In this way, the AOI equipment can comprehensively and accurately analyze the overall image of the chip, detect the multi-wire and misaligned wires of the chip leads, and the integrity of the chip body, thereby ensuring the quality of the chip.

[0104] In one embodiment, based on the above embodiment, after the step of marking non-defective chips as qualified products, marking defective chips as unqualified products, marking chips that have failed defect detection as pending products, and generating chip defect detection results, the method further includes: Correlate the defect detection results corresponding to multiple batches of chips with the corresponding bonding parameters and array positions to generate analysis samples; wherein the array arrangement during chip defect detection is the same as the array arrangement during gold wire bonding; The analysis samples are input into a pre-trained artificial intelligence model for defect cause analysis to obtain a cause report for each type of defect; wherein, the training samples used in training the artificial intelligence model include analysis samples of defective chips and analysis samples of non-defective chips.

[0105] In this embodiment, in order to conduct in-depth defect cause analysis, it is necessary to associate the defect detection results with relevant manufacturing parameters and position information. Among them, bonding parameters include but are not limited to bonding pressure, temperature, time, etc. These parameters are directly related to the quality of bonding and are key factors that may cause defects; the array position is the specific position of the chip in the detection area and the position during the gold wire bonding process. Since the chips are processed in batches, different positions may be affected by the uniformity of the equipment, so the position information is very important for analyzing the distribution pattern of defects.

[0106] The defect detection results are matched with the corresponding bonding parameters and array position information to generate analysis samples, where each analysis sample contains the following information: chip ID, defect detection results, bonding parameters, and array position.

[0107] Among them, it is ensured that the arrangement array during chip defect detection is the same as the arrangement array during gold wire bonding, so that the defects can be accurately associated with specific manufacturing conditions and analysis errors caused by position mismatch can be avoided.

[0108] In order to perform defect cause analysis, a trained artificial intelligence model is required. This model can learn from analysis samples and identify which bonding parameters and location factors are associated with specific types of defects. Among them, the training samples should include two types of data: one is the analysis samples of defective chips, which contain known defects and their corresponding bonding parameters and location information to help the model learn the characteristics of defects; the other is the analysis samples of non-defective chips, which serve as references for normal chips to help the model distinguish between normal and abnormal situations and improve the accuracy of predictions.

[0109] Optionally, various machine learning or deep learning models can be selected for model selection, such as decision trees, random forests, support vector machines, neural networks, etc. The specific selection depends on the characteristics of the data and the needs of the analysis. Then determine which bonding parameters and position information are the features that have the greatest impact on defects. This can be achieved through feature selection algorithms to improve the efficiency and accuracy of the model. The specific model training process is as follows: (1) Data preparation First, a large amount of historical data needs to be collected, including analysis samples of defective chips and analysis samples of non-defective chips. The collected data is cleaned to handle missing values, outliers, and inconsistent data.

[0110] Useful features are then extracted from the raw data, which can include bonding parameters and chip location information. In addition, interactive features can be created if the combination of certain parameters may have a greater impact on defects.

[0111] The dataset is divided into training set, validation set and test set, with optional ratios of 70%, 15%, 15% or 80%, 10%, 10%.

[0112] (2) Model selection Select an appropriate machine learning or deep learning model. For defect cause analysis, the random forest model is preferred. The random forest model is suitable for multi-classification problems, can handle high-dimensional data, and has a certain degree of interpretability.

[0113] (3) Model training The selected model is trained using the training set data. In this process, the model learns the relationship between features and defect types. Also, the validation set is used to tune the model's hyperparameters to achieve the best performance. Methods such as grid search, random search, or Bayesian optimization can be used.

[0114] (4) Model evaluation Use the test set to evaluate the performance of the model. Common evaluation indicators include: Accuracy, Precision, Recall, F1 score and confusion matrix (intuitively display the classification of the model).

[0115] (5) Model deployment Deploy trained models to production environments for real-time or batch defect cause analysis. As new data is collected, models can be updated regularly to maintain their performance. In addition, they can be integrated with existing production systems and databases to enable automatic data flow.

[0116] Optionally, the generated analysis samples are input into the trained AI model, which will analyze the causes of defects based on the learned patterns and relationships and generate a report on the causes of each type of defect. The report will indicate which bonding parameters or position factors each defect type may be associated with, and how the interaction between these factors affects the occurrence of defects. In order for engineers to understand the analysis results of the model, visual charts or explanatory text can be provided to clearly show the relationship between defects and parameters.

[0117] In one embodiment, based on defect cause analysis, bonding process parameters can be adjusted in a targeted manner, the production process can be optimized, and the defect rate can be reduced; by analyzing the distribution and trend of defects, potential problems of equipment can be predicted, preventive maintenance can be performed, and downtime can be reduced. This can reduce the proportion of defective chips, directly reduce production costs, and improve production efficiency.

[0118] In this way, by using big data and artificial intelligence technology to achieve data-driven decision-making and enhance the intelligence level of the entire manufacturing process, not only can the efficiency and accuracy of defect detection be improved, but also through in-depth analysis of the causes of defects, continuous improvement and optimization of the production process can be promoted.

[0119] In one embodiment, based on the above embodiment, after the step of inputting the analysis sample into the pre-trained artificial intelligence model to perform defect cause analysis to obtain a cause report of each type of defect, the step further includes: Based on the cause report, the artificial intelligence model is used to further analyze the optimal bonding parameters corresponding to each array position; Based on the optimal bonding parameters, the bonding parameters selected by the chip bonding equipment associated with the AOI equipment are updated.

[0120] In this embodiment, in addition to pre-training the defect cause analysis model, the artificial intelligence model can also simultaneously train the optimal bonding parameter analysis model.

[0121] Based on the defect cause analysis model training, the optimal bonding parameter combination can be searched by defining the optimization goal (such as minimizing the defect rate or maximizing the bonding quality) and selecting a suitable optimization model (such as genetic algorithm, particle swarm optimization, Bayesian optimization, etc.).

[0122] Optionally, an optimization algorithm is used to search for the best bonding parameters for each array position. For each position, the model tries different combinations of parameters to find the best solution that minimizes the defect rate.

[0123] After obtaining the cause report of each type of defect, the AI ​​model can further analyze the optimal bonding parameters corresponding to different array positions. The purpose of this step is to identify which bonding parameter settings can minimize the occurrence of defects at different positions.

[0124] Optionally, analyze the relationship between different array positions and bonding parameters to find out which parameter combinations at specific positions lead to defects. For each defect type, determine which bonding parameters are the main influencing factors.

[0125] Optionally, an optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) is used to search for the best bonding parameter combination for each array position. If necessary, a physical model or simulation tool can also be used to predict the bonding quality under different parameter settings to assist in determining the optimal parameters.

[0126] Through the above analysis, a set of optimal bonding parameters are determined for each array position. These parameters are designed to minimize the defect rate and improve the bonding quality.

[0127] Generate a mapping table that lists the optimal bonding parameters for each array position in detail. Provide specific parameter adjustment suggestions, such as setting values ​​for bonding pressure, temperature, time, etc.

[0128] Optionally, select some array positions in actual production, perform bonding according to the new parameter settings, and monitor whether the defect rate is improved. Verify the effectiveness of the new parameter settings. Once the optimal bonding parameters are determined and their effectiveness is verified through experiments, these parameters can be updated to the chip bonding equipment associated with the AOI equipment.

[0129] Optionally, automatic synchronization of parameters between the AI ​​model and the bonding equipment can be achieved through system integration. Continuous monitoring of the bonding quality after the updated parameters ensures that the defect rate is effectively controlled. Regularly re-evaluate and update the optimal parameters to adapt to changes in the production process.

[0130] It should be noted that during the operation of the chip bonding equipment, due to mechanical wear, material deformation or other physical factors, the actual effects of parameters such as bonding pressure, temperature and time at different locations may be different. For example, the bonding head at certain locations may be more severely worn, or the mechanical structure of the equipment may be loose in certain areas. In addition, during the bonding process, the temperature distribution of the heating area may be uneven. Even if an average temperature is set, chips at different locations may receive different temperatures, which will affect the bonding effect.

[0131] Through data analysis and optimization algorithms, the optimal bonding parameters can be customized for each array position to minimize the occurrence of defects, thereby reducing rework and scrap caused by defects and reducing production costs. The entire process is based on data analysis and artificial intelligence technology to achieve intelligent management of the production process.

[0132] In this way, by applying the analysis results of the artificial intelligence model to the optimization of bonding parameters and updating them to the production equipment in real time, a closed-loop production optimization system can be formed to continuously improve the quality and efficiency of chip manufacturing.

[0133] In one embodiment, based on the above embodiment, after the step of updating the bonding parameters selected by the chip bonding equipment associated with the AOI equipment based on the optimal bonding parameters, the step further includes: After the bonding parameters of the chip bonding equipment are updated, if the defect detection results of the latest batch of chips are improved, the model parameters of the artificial intelligence model are updated using the analysis samples corresponding to the latest batch of chips; Based on the updated AI model, continue to optimize bonding parameters.

[0134] In this embodiment, after the bonding parameters of the chip bonding equipment are updated, the latest batch of chips is tested for defects and the test results are recorded. The defect detection results of the new batch of chips are compared with the old batch of chips to evaluate the effect of the updated bonding parameters. If the defect detection results of the latest batch of chips are improved (for example, the number of unqualified chips is reduced), proceed to the next step.

[0135] Collect the bonding parameters, array position information, and defect detection results of the latest batch of chips. After cleaning and annotating the new sample data, use the new sample data to retrain or incrementally train the existing artificial intelligence model to update the model parameters.

[0136] Generate defect cause report using updated AI model, and continue to search for optimal bonding parameter combination for each array position using optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) based on the latest cause report. List the latest optimal bonding parameters corresponding to each array position in detail, and update the new optimal bonding parameters to the bonding equipment.

[0137] In actual production testing, some array positions can be selected, bonded according to the new parameter settings, and monitored to see if the defect rate is further improved. The optimal parameters are regularly re-evaluated and updated to adapt to changes in the production process.

[0138] Through the above steps, a closed-loop system with continuous improvement can be realized. After each bonding parameter update, the latest defect detection results are evaluated to decide whether to further update the artificial intelligence model. In this way, the model can continuously learn new data, optimize bonding parameters, and further improve the quality and efficiency of chip manufacturing. The entire process is based on data analysis and artificial intelligence technology to achieve intelligent management of the production process.

[0139] In one embodiment, based on the above embodiment, after the step of marking non-defective chips as qualified products, marking defective chips as unqualified products, marking chips that have failed defect detection as pending products, and generating chip defect detection results, the method further includes: Qualified products, unqualified products and products to be confirmed are sorted, and qualified products are transported to the downstream equipment of the AOI equipment or stored in the material box, unqualified products are transported to the recycling area, and products to be confirmed are transported to the manual inspection area.

[0140] In this embodiment, during the defect detection process, the specific position of each chip in the array is recorded (for example, represented by a coordinate system), and the defect detection result of each chip is associated with its position information in the array to form a mapping table including position information and detection results.

[0141] The conveyor belt in the AOI equipment starts working, moving the inspected chips from the inspection platform to the sorting area. The loading and unloading mechanism is ready to receive the chips on the conveyor belt and transfer them to the working range of the robot arm. The robot arm reads the position information of the chip in the array, and can move the chips marked as "qualified products" to the conveyor belt of the downstream equipment according to the position information and the results in the mapping table, and continue the subsequent processes (such as packaging, testing, etc.), or store the qualified chips in the preset material box for subsequent use.

[0142] A robotic arm moves chips marked as "rejected" to a conveyor belt in the recycling area, where they will be sent to recycling or scrap processing areas.

[0143] The robotic arm moves the chips marked as "pending confirmation" to the conveyor belt in the manual inspection area, where they will be further inspected and confirmed by staff.

[0144] The AOI equipment records the sorting results of each chip, including the sorting time, sorting location, and sorting destination (downstream equipment, material boxes, recycling areas, manual inspection areas).

[0145] Through the above steps, AOI equipment can realize automatic classification and sorting of chips. Such an automated sorting system can ensure efficient and accurate classification of chips, improve production efficiency, reduce manual intervention and errors, and ensure that chip quality management is more accurate and efficient.

[0146] In one embodiment, based on the above embodiment, the multi-angle defect detection method for batch chips further includes: When taking panoramic images of the chip array in all directions, the relative position relationship between the camera array and the chip array is obtained based on visual detection; According to the relative position relationship between the camera array and the chip array, the initial focal length of each camera in the camera array is uniformly set; According to the array position of each camera and the light-sensing data detected by each camera based on the initial focal length, the shooting area that each camera is responsible for is determined; wherein the set of shooting areas that each camera is responsible for covers the area where the chip array is located; Based on the relative positional relationship between the camera array and the chip array, the relative positional relationship between each camera and the shooting area it is responsible for is obtained based on visual detection; Based on the relative positional relationship between each camera and the shooting area it is responsible for, and the light-sensitive data detected by each camera, the focal length of each camera is adjusted on the basis of the initial focal length.

[0147] In this embodiment, before taking the panoramic image of the chip array, the camera array obtains the relative position relationship between the camera array and the chip array through visual detection technology.

[0148] The parameter description of the relative position relationship may include the distance, angle, offset, etc. between the camera array and the chip array. The distance is the spatial distance between the camera array and the chip array, the angle is the angular deviation between the camera array and the chip array (such as horizontal angle, vertical angle), and the offset is the geometric center offset of the camera array relative to the chip array.

[0149] Since the camera array has multiple cameras, it is possible to calculate the distance and angle between the camera array and the shooting area based on multi-view visual detection through stereo vision technology, and combine image processing algorithms to calculate the spatial relationship between the two. The system will generate an accurate relative position data for subsequent focus setting and shooting area division.

[0150] After obtaining the relative position relationship between the camera array and the chip array, calculate the appropriate focal length value according to the distance between the camera array and the chip array, and uniformly set the initial focal length of each camera in the camera array. Ensure that the focal length setting of all cameras is reasonable when starting to shoot, to avoid image blur or out-of-focus problems caused by focal length differences.

[0151] For example, if the distance between the camera array and the chip array is 10 cm, a suitable focal length value (such as 50 mm) is calculated and uniformly set to all cameras.

[0152] A camera array is a set of multiple cameras arranged in a certain layout (such as a linear array or a matrix array). The array position of each camera determines its position and angle in the entire detection system. Based on the array position of each camera and its photosensitivity data (such as brightness, contrast, etc.) at the initial focal length, the specific shooting area that each camera is responsible for is determined.

[0153] Among them, the photosensitive data can help determine whether each camera can clearly capture the area it is responsible for at the initial focal length.

[0154] Area coverage requirement: The shooting area set of all cameras must cover the area where the entire chip array is located, ensuring that each chip can be photographed by at least one camera.

[0155] For example, if the camera array is arranged linearly, camera 1 is responsible for the left area of ​​the chip array, camera 2 is responsible for the middle area, and camera 3 is responsible for the right area; in a matrix array, camera 1 is responsible for the upper left area, camera 2 is responsible for the upper right area, and so on.

[0156] The previously acquired relative position relationship between the camera array and the chip array can provide a global reference for the subsequent relative position detection between each camera and its shooting area. Including: A camera array is usually composed of multiple cameras arranged in a certain layout (such as a linear array, matrix array, or ring array). The array position of each camera determines its position and angle in the overall system.

[0157] Each camera needs to obtain the specific relative position relationship between it and the shooting area it is responsible for through visual detection to ensure that it can accurately focus and clearly shoot the area it is responsible for. Among them, image matching technology (such as SIFT, ORB, etc.) can be used to match the image taken by the camera with the pre-stored chip array template to calculate the relative position.

[0158] Each camera transmits the relative position data it detects to the system, and the system integrates the relative position data of all cameras to generate a global position relationship diagram to ensure that each camera clearly knows its relative position to the shooting area.

[0159] Example scenario: Example 1: The distance between camera 1 and the shooting area it is responsible for is 12 cm, the horizontal angle deviation is 2 degrees, the vertical angle deviation is 1 degree, and the geometric center offset is (0.5 mm, 0 mm); Example 2: The distance between camera 2 and the shooting area it is responsible for is 10 cm, the horizontal angle deviation is 0 degrees, the vertical angle deviation is 0 degrees, and the geometric center offset is (0 mm, 0 mm); Example 3: The distance between camera 3 and the shooting area it is responsible for is 11 cm, the horizontal angle deviation is -1 degree, the vertical angle deviation is 0.5 degree, and the geometric center offset is (0 mm, -0.3 mm).

[0160] Optionally, according to the precise relative position relationship between each camera and its shooting area, and the detected light-sensitive data, fine-tuning is performed on the basis of the initial focal length to ensure that each camera can clearly capture the shooting area it is responsible for.

[0161] Optionally, the focal length value that needs to be adjusted is calculated based on the distance and angle deviation between each camera and its shooting area. Combined with the light-sensitive data (such as image clarity, contrast, etc.), the effect of the focal length adjustment is judged to further optimize the focal length setting.

[0162] For example, if a camera is far away from its shooting area, the focal length is appropriately increased to ensure a clear image; if a camera is close to its shooting area, the focal length is appropriately reduced to avoid image distortion. Through cyclic adjustment and detection, it is ensured that the focal length of each camera is suitable for its shooting area.

[0163] In this way, it can be ensured that every part of the chip array is clearly photographed, and at the same time, image blur or distortion can be avoided by adjusting the focus, so as to provide high-quality image data for subsequent defect detection and improve the accuracy and efficiency of detection.

[0164] In addition, a control device is also provided in an embodiment of the present application. The internal architecture of the control device can be as follows: Figure 3 As shown, it includes a processor, a memory, a communication interface and an input interface connected by a system bus. Among them, the processor is used to provide computing and control capabilities. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database is used to store data called by the computer program. The communication interface is used to communicate data with an external terminal. The input interface is used to receive a signal input by an external device. When the computer program is executed by the processor, a multi-angle defect detection method for batch chips as described in the above embodiment is implemented.

[0165] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the control device to which the solution of the present application is applied. For example, in some optional embodiments, the control device may also include an output interface (not shown in the figure), and the output interface is also connected to the system bus and is used to output corresponding signals to the peripheral device.

[0166] In addition, the present application also proposes an AOI device, which includes a camera array, a motion camera, and a control device; wherein the camera array and the motion camera both establish a communication connection with the control device and are controlled by the control device; The control device is the control device as described above.

[0167] In addition, the present application also proposes a computer-readable storage medium, the computer-readable storage medium includes a computer program, and when the computer program is executed by a processor, the steps of the batch chip multi-angle defect detection method as described in the above embodiment are implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0168] In summary, the multi-angle defect detection method, control device, AOI device and computer-readable storage medium for batch chips provided in the embodiments of the present application place the chip array after gold wire bonding into the detection area of ​​the AOI device, and use the camera array to simultaneously shoot multiple angles of multiple chips, which significantly reduces the independent shooting time of each chip and improves the overall detection efficiency. While performing chip defect detection on qualified images, the motion camera can also be used to intelligently reshoot some images of insufficient quality taken by the camera array to ensure the quality and efficiency of detection. In addition, during the reshooting process, unnecessary movements and pauses are reduced by intelligently planning the path and optimizing the motion control of the camera, which not only takes into account the reshooting efficiency and the quality of the reshooting images, but also reduces mechanical wear and extends the service life of the equipment. In this way, it helps to improve the overall defect detection efficiency of batch chips, thereby improving the production efficiency of batch chips and ensuring chip quality.

[0169] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0170] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0171] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A multi-angle defect detection method for batch chips, characterized in that: include: The chip array after gold wire bonding is placed in the detection area of ​​the AOI equipment; wherein the detection area is provided with a camera array and a motion camera; Control the camera array to operate at multiple shooting angles to capture panoramic images of the chip array in multiple directions; during the camera array shooting process, each camera focuses independently; Perform image segmentation on panoramic images in multiple directions to obtain multi-angle images corresponding to each chip; Perform image quality checks on images from multiple angles and filter out unqualified images; For chips without unqualified images, multi-angle images are stitched to obtain the overall image of the chip, and the overall image of the chip is analyzed based on the AOI technology to detect defects on the chip; and the image reshoot path is intelligently planned according to the chip position and orientation to which the unqualified image belongs; the path in the same orientation is calculated according to the shortest distance of the connecting line of the chip positions to which all unqualified images in the same orientation belong; and the paths in different orientations are connected to form the image reshoot path according to the shortest switching distance between the paths in different orientations; Control the motion camera to retake the unqualified images along the image retake path; wherein the interval path of the motion camera between the two retake points includes a speed-up segment, a speed-down segment, a uniform speed preparation segment and a uniform speed capture segment in sequence, and the length of the uniform speed preparation segment is positively correlated with the deceleration of the speed-down segment; the end point of the uniform speed capture segment of the previous interval path is the start point of the speed-up segment of the next interval path; Replace unqualified images with retaken images; For the chip after the image reshoot, multi-angle images are stitched together to obtain the overall image of the chip, and the overall image of the chip is analyzed based on AOI technology to detect defects on the chip; Mark non-defective chips as qualified products, mark defective chips as unqualified products, mark chips that fail to complete defect detection as pending confirmation products, and generate defect detection results for the chips.

2. The multi-angle defect detection method for batch chips according to claim 1, characterized in that: The step of performing image analysis on the overall image of the chip based on the AOI technology to detect defects on the chip includes: Locate the gold wire image in the overall chip image based on the edge detection algorithm and analyze whether the wiring in the gold wire image is correct; Based on image segmentation technology, the chip body image is separated from the overall chip image, and the chip body image is detected to see if there are defects.

3. The multi-angle defect detection method for batch chips according to claim 1 or 2, characterized in that: After the step of marking the non-defective chips as qualified products, marking the defective chips as unqualified products, marking the chips that failed to complete the defect detection as pending products, and generating the defect detection results of the chips, the method further includes: Correlate the defect detection results corresponding to multiple batches of chips with the corresponding bonding parameters and array positions to generate analysis samples; wherein the array arrangement during chip defect detection is the same as the array arrangement during gold wire bonding; The analysis samples are input into a pre-trained artificial intelligence model for defect cause analysis to obtain a cause report for each type of defect; wherein, the training samples used in training the artificial intelligence model include analysis samples of defective chips and analysis samples of non-defective chips.

4. The multi-angle defect detection method for batch chips as claimed in claim 3, characterized in that: After the step of inputting the analysis sample into the pre-trained artificial intelligence model to perform defect cause analysis to obtain a cause report of each type of defect, the method further includes: Based on the cause report, the artificial intelligence model is used to further analyze the optimal bonding parameters corresponding to each array position; Based on the optimal bonding parameters, the bonding parameters selected by the chip bonding equipment associated with the AOI equipment are updated.

5. The multi-angle defect detection method for batch chips as claimed in claim 4, characterized in that: After the step of updating the bonding parameters selected by the chip bonding equipment associated with the AOI equipment based on the optimal bonding parameters, the method further includes: After the bonding parameters of the chip bonding equipment are updated, if the defect detection results of the latest batch of chips are improved, the model parameters of the artificial intelligence model are updated using the analysis samples corresponding to the latest batch of chips; Based on the updated AI model, continue to optimize bonding parameters.

6. The multi-angle defect detection method for batch chips as claimed in claim 1, characterized in that: After the step of marking the non-defective chips as qualified products, marking the defective chips as unqualified products, marking the chips that failed to complete the defect detection as pending products, and generating the defect detection results of the chips, the method further includes: Qualified products, unqualified products and products to be confirmed are sorted, and qualified products are transported to the downstream equipment of the AOI equipment or stored in the material box, unqualified products are transported to the recycling area, and products to be confirmed are transported to the manual inspection area.

7. The multi-angle defect detection method for batch chips as claimed in claim 1, characterized in that: The multi-angle defect detection method for batch chips also includes: When taking panoramic images of the chip array in all directions, the relative position relationship between the camera array and the chip array is obtained based on visual detection; According to the relative position relationship between the camera array and the chip array, the initial focal length of each camera in the camera array is uniformly set; According to the array position of each camera and the light-sensing data detected by each camera based on the initial focal length, the shooting area that each camera is responsible for is determined; wherein the set of shooting areas that each camera is responsible for covers the area where the chip array is located; Based on the relative positional relationship between the camera array and the chip array, the relative positional relationship between each camera and the shooting area it is responsible for is obtained based on visual detection; Based on the relative position relationship between each camera and the shooting area it is responsible for, and the light-sensitive data detected by each camera, the focal length of each camera is adjusted on the basis of the initial focal length.

8. A control device, characterized in that: The control device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of the multi-angle defect detection method for batch chips as described in any one of claims 1 to 7 are implemented.

9. An AOI device, characterized in that: The AOI device includes a camera array, a motion camera and a control device; wherein the camera array and the motion camera are both connected to the control device in communication and are controlled by the control device; The control device is the control device according to claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the multi-angle defect detection method for batch chips according to any one of claims 1 to 7 are implemented.

Images