Automatic dust removal method and equipment for batch chips, dust removal detector and medium

Abstract

The invention relates to an automatic cleaning technology, and discloses an automatic dust removal method and equipment for batch chips, a dust removal detector and a medium, and the method comprises the following steps: placing a plurality of cleaned chips into a detection area of the dust removal detector; controlling the motion camera to shoot images of the chips in the detection area one by one; synthesizing a chip image when all azimuth images of the same chip are shot; when the chip image of each chip is obtained, performing stain detection on the chip image by using an image recognition technology; if the chip is detected to have the stain, recording a stain coordinate according to the chip position where the stain is located, and associating the stain coordinate with the chip area where the stain is located and the stain information; after stain detection is completed on all the chips, a dust removal track is generated; and controlling the manipulator to clean the chip stains at all positions one by one by using a dust removal rod along the dust removal track. The invention aims to realize efficient and accurate automatic dust removal of batch chips.

Description

Technical Field

[0001] The present application relates to the field of automatic cleaning technology, and in particular to an automated dust removal method for batch chips, a control device, a dust removal detection machine, and a computer-readable storage medium. Background Art

[0002] In the semiconductor manufacturing industry, the cleanliness of the chip has a crucial impact on the performance and reliability of the product. With the advancement of technology, the size of the chip continues to shrink, while its function becomes increasingly powerful, which means that any tiny stain on the surface of the chip may have a significant impact on its performance. Take the photosensitive chip of the camera module as an example. If you want to ensure that the camera module has excellent imaging effects, you must ensure that the photosensitive chip area is highly clean, dust-free, and dirt-free.

[0003] During the chip production process, although cleaning equipment such as centrifugal cleaning machines and deionized water washing machines are used to clean the chips, even so, after cleaning, some dirt that is difficult to remove with general cleaning equipment will still remain (for example, centrifugal cleaning machines are better at removing larger particles, but it is difficult to completely remove tiny particles, especially dirt embedded on the chip surface; deionized water washing machines have limited cleaning effects on some insoluble or strongly adherent dirt, such as organic residues; there may be water stains on the surface of the chip after cleaning, and these water stains may form water marks during the drying process, affecting the surface finish of the chip), resulting in defective products and then causing product scrapping.

[0004] 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

[0005] The main purpose of this application is to provide an automated dust removal method for batch chips, a control device, a dust removal detection machine and a computer-readable storage medium, aiming to achieve efficient and accurate automated dust removal for batch chips.

[0006] To achieve the above objectives, the present application provides an automated dust removal method for batch chips, comprising the following steps: Place the cleaned chips into the detection area of ​​the dust removal detection machine; Controlling the motion camera to take images of the chips in the detection area one by one; wherein the motion camera takes images of each direction of each chip one by one, and after taking images of all directions of the same chip, takes images of the next chip; After all the orientation images of the same chip are captured, all the orientation images of the same chip are synthesized to obtain a chip image of the same chip; Each time a chip image of a chip is obtained, the chip image is subjected to stain detection using image recognition technology; If stains are detected on the chip, the stain coordinates are recorded according to the chip position where the stain is located, and the stain coordinates are associated with the chip area where the stain is located and the stain information, wherein the stain information includes the area size and color depth; and the robot is controlled to go to the dust removal rod placement area to clamp the dust removal rod and move it to the standby position; After completing the stain detection on all chips, the robot's standby position is used as the starting point to connect all the stain coordinates in sequence, and the shortest path planning is performed to generate the dust removal trajectory; The robot is controlled to use a dust removal rod to remove chip stains one by one along a dust removal track; wherein, according to the force upper limit value and stain information of the chip area where each stain is located, the pressing force of the robot when using the dust removal rod to wipe the stain is adjusted.

[0007] 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 automated dust removal method for batch chips as described above are implemented.

[0008] To achieve the above-mentioned purpose, the present application also provides a dust removal detection machine, which includes a control device, a motion camera and a manipulator, and the motion camera and the manipulator are both connected to the control device in communication and are controlled by the control device; the control device is the control device as described above.

[0009] 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 automated dust removal method for batch chips as described above are implemented.

[0010] The automated dust removal method, control device, dust removal detection machine and computer-readable storage medium for batch chips provided by this application realize efficient and accurate automated dust removal for batch chips by combining image recognition, automated mechanical operation and shortest path planning, which not only improves the cleanliness of chips and reduces the production of defective products, but also improves production efficiency and reduces production costs. This has important application value for the semiconductor manufacturing industry, especially for products such as camera modules that have extremely high cleanliness requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a schematic diagram of the steps of an automated dust removal method for batch chips in one embodiment of the present application; Figure 2 Schematic diagram of the internal structure of a control device according to an embodiment of the present application.

[0012] 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

[0013] 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.

[0014] 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.

[0015] Reference Figure 1 In one embodiment, the automated dust removal method for batch chips includes: Step S10, placing the cleaned chips into the detection area of ​​the dust removal detection machine; Step S20, controlling the motion camera to capture images of the chips in the detection area one by one; wherein the motion camera captures images of each chip in each direction one by one, and after capturing images of all directions of the same chip, it captures images of the next chip; Step S30, after all the orientation images of the same chip are obtained by shooting, all the orientation images of the same chip are synthesized to obtain a chip image of the same chip; Step S40: After obtaining a chip image of a chip, use image recognition technology to perform stain detection on the chip image; Step S50: if stains are detected on the chip, the stain coordinates are recorded according to the chip position where the stains are located, and the stain coordinates are associated with the chip area where the stains are located and the stain information, wherein the stain information includes the area size and color depth; and the manipulator is controlled to go to the dust removal rod placement area to clamp the dust removal rod and move to the standby position; Step S60: After all chips have been contaminated, the standby position of the robot is used as the starting point, all contaminated coordinates are connected in sequence, and the shortest path planning is performed to generate a dust removal trajectory; Step S70, control the robot arm to remove chip stains one by one using a dust removal rod along the dust removal track; wherein, according to the force upper limit value and stain information of the chip area where each stain is located, adjust the pressing force of the robot arm when using the dust removal rod to remove the stain.

[0016] In this embodiment, the execution terminal of the embodiment may be a control device, or may be other equipment or device that controls the control device.

[0017] As described in step S10, after the batch-produced chips are cleaned by general cleaning equipment (such as a centrifugal cleaning machine, a deionized water washing machine), there may still be some tiny stains on the chip surface, which need to be further processed by automated dust removal equipment.

[0018] Optionally, the cleaned chips can be transferred from the cleaning equipment to the detection area of ​​the dust removal inspection machine via a conveyor belt or other automated conveying device. During the transmission process, it is necessary to ensure that the direction and position of the chips are consistent to avoid collision or stacking between chips. Among them, the operator can place a material box loaded with multiple chips produced in the same batch into the relevant equipment, and then push each tray of chips from the material box to the conveyor belt through the loading mechanism, and then the conveyor belt will transfer the chips to the detection area of ​​the dust removal inspection machine.

[0019] Optionally, a lifting platform may be provided at the exit of the conveyor belt in the detection area, and the lifting platform may be provided with multiple chip placement positions for receiving each chip to be detected and dusted. The chips may be placed one by one at the corresponding position of the lifting platform by a manipulator or a worker wearing dustproof gloves; each chip placement position of the lifting platform may also be provided with a vacuum air hole to adsorb and fix each chip; the vacuum air hole provides negative pressure through a vacuum pump to firmly adsorb the chip at the placement position, ensuring that the chip will not shift during the detection and dust removal process.

[0020] Optionally, a batch of chips can be lifted to a shooting position suitable for action camera shooting by a lifting platform.

[0021] The core task of step S10 is to place the cleaned chips safely and neatly into the detection area of ​​the dust removal inspection machine to prepare for subsequent image capture and stain detection. Through precise positioning, clean environment control, and automated or semi-automated chip placement, the stability and reliability of the chips during the detection and dust removal process can be ensured, thereby improving production efficiency and product quality.

[0022] As described in step S20, a motion camera is used to take omnidirectional images of each chip in the detection area for subsequent stain detection.

[0023] Motion cameras can use high-resolution industrial cameras to capture tiny details on the chip surface. The camera's pixels and frame rate can be selected based on the chip size and the accuracy requirements for stain detection; motion cameras can achieve multi-angle shooting through a motion control platform (such as an XY platform, a mechanical device that can move accurately in two mutually perpendicular directions (X and Y)).

[0024] In the detection area, the chip has been accurately placed at a fixed position through step S10. The motion camera can capture images of the chips one by one according to a preset shooting sequence.

[0025] According to the order in which the chips are arranged in the detection area, the motion camera will first shoot the first chip, and then move to the next chip. The control of the shooting order is completed by the control device to ensure that no chip is missed.

[0026] The motion camera needs to capture images of each chip from multiple angles, including the front and four sides. The number and position of the shooting angles are optimized according to the size and shape of the chip to ensure that every area of ​​the chip surface is covered.

[0027] After the action camera completes shooting in one direction (such as the front), it adjusts its position through the operation control platform and moves to the next direction (such as the side) to shoot. After all the orientation images of each chip are shot, the action camera will move to the next chip.

[0028] The movement speed of the action camera can be adjusted according to the size of the chip and the frame rate of the camera to ensure that the shooting process is efficient and stable.

[0029] The motion camera collects image data in real time during the shooting process and transmits it to the control system through a data cable. The image data after shooting is stored in the high-speed storage device of the control system. A solid-state hard disk can be used to increase the reading and writing speed. The image files are named and classified according to the chip number and shooting direction to facilitate subsequent stain detection and analysis.

[0030] The light source in the detection area needs to be stable, and the brightness, angle, and color temperature of the light source need to match the imaging characteristics of the motion camera. Among them, a uniform LED light source can be used to avoid image shadows or overexposure caused by uneven light sources.

[0031] In step S20, a motion camera is used to take images of each chip in the detection area one by one and at multiple angles to ensure that stains on the surface of the chip can be accurately captured.

[0032] As described in step S30, for each chip, images of the chip in different directions are synthesized to generate a complete chip image, so as to facilitate subsequent stain detection and analysis.

[0033] The multi-directional image data of each chip captured in step S20 is read from the storage device. Since there may be a slight position deviation during the shooting process, the images in different directions need to be aligned to ensure that the relative positions between the images are consistent.

[0034] Among them, feature point matching or global alignment algorithm is used to align images in different orientations to the same coordinate system.

[0035] Optionally, the captured image is subjected to denoising to reduce noise interference caused by the shooting environment or equipment. A filter (such as a Gaussian filter or a median filter) is used to smooth the image.

[0036] Perform geometric correction on the image to correct the image deformation caused by lens distortion or shooting angle, such as using a distortion correction algorithm (such as the undistort function in OpenCV) to correct the image.

[0037] Optionally, use a panoramic stitching algorithm (such as the OpenCV library in the image stitching tool) to stitch images from different directions into a panoramic image. During the stitching process, the overlapping areas of the images need to be fused to ensure a smooth transition at the stitching point.

[0038] Alternatively, images from different perspectives are fused to generate an image that comprehensively reflects the surface conditions of the chip. Methods such as weighted average and maximum value fusion can be used to ensure that the synthesized image is rich in details and has no obvious splicing marks.

[0039] Optionally, the composite image is contrast enhanced to improve the visual effect of the image and make details such as stains more obvious. Among them, the enhancement can be performed using methods such as histogram equalization and adaptive histogram equalization.

[0040] Optionally, a color correction algorithm is used to adjust the RGB components of the image to correct the color of the composite image to ensure that the hue and brightness of the image are uniform and consistent.

[0041] Step S30 synthesizes all orientation images of the same chip to generate an image that comprehensively reflects the surface conditions of the chip, thereby providing high-quality image data for subsequent stain detection and analysis.

[0042] As described in step S40, for each chip image generated in step S30, the stains on the chip surface are detected and analyzed using image recognition technology. This process is intended to identify whether there are stains on the chip surface, and accurately locate and classify their position, size and type, providing a basis for subsequent stain treatment.

[0043] The synthesized chip image is obtained from step S30 as input data for stain detection.

[0044] Optionally, the color image is converted to a grayscale image to simplify the image processing process and reduce the computational complexity.

[0045] Optionally, the grayscale image is binarized to convert the image into a black and white binary image to facilitate the extraction of the stain area. The binarization threshold is determined using a threshold segmentation algorithm (such as Otsu algorithm, adaptive threshold).

[0046] Perform edge detection on the grayscale image to extract the edge information of the chip surface structure to provide a reference for stain detection. The Canny edge detection algorithm is used to identify edge features in the chip image.

[0047] Optionally, the grayscale image or the binary image is enhanced to improve the contrast and visibility of the stain area, wherein the image details are enhanced by using methods such as histogram equalization and sharpening filters.

[0048] Optionally, a morphological operation (such as erosion, dilation, opening, and closing) is used to process the binary image to extract the stain area. Morphological operations can remove noise from the image while preserving the shape and size of the stain.

[0049] Optionally, a region growing algorithm is used to select seed points from the image, and regions similar to the seed points are gradually grown to identify the stain region. The region growing algorithm is suitable for detecting stains with irregular shapes.

[0050] Optionally, a connected region analysis is performed on the binary image to identify and mark the connected regions (i.e., possible stain regions) in the image, wherein a four-connected or eight-connected algorithm is used to mark the stain regions and calculate their areas and positions.

[0051] Alternatively, a pre-trained convolutional neural network model is used to extract features and detect stains on the chip image; alternatively, a target detection algorithm is used to locate and classify stains in the chip image. The target detection algorithm can simultaneously identify multiple stains and provide their location and category information.

[0052] Optionally, use an image segmentation algorithm (such as U-Net, Mask R-CNN) to accurately segment the stain area in the chip image. The image segmentation algorithm can generate pixel-level annotations of the stain area to improve detection accuracy.

[0053] Optionally, a suitable detection threshold is set according to the characteristics of the stain on the chip surface to ensure the sensitivity and specificity of the detection.

[0054] Optionally, the detected stain area is marked on the chip image, indicating its location, size and type. The stain area can be marked with a different color or a border to facilitate subsequent visual inspection and processing.

[0055] Optionally, the stains may be classified into different types, such as particulate matter, liquid residue, oxide, etc., according to their characteristics (e.g., color, shape, material).

[0056] Step S40 performs stain detection on the chip image through image recognition technology, accurately identifies the stain area on the chip surface, and analyzes its position, size and type.

[0057] As described in step S50, for the chip detected to have stains in step S40, it is necessary to accurately record the location information of the stains and associate it with the area information of the chip, and at the same time control the robot arm to prepare for dust removal operation.

[0058] Record the chip with stains, and use the specific position of the chip in the detection area as the stain coordinates. Each chip has a specific position in the detection area, which can be represented by a coordinate system. At the same time, based on a reference point in the detection area (such as the lower left corner or the center point), record the specific position coordinates of the chip. This position coordinate will be used as a reference point for subsequent stain coordinates to record the chip area where the stain is located.

[0059] Optionally, the stain coordinates are associated with the chip region where the stain is located, the size of the area, and the color depth.

[0060] Optionally, the chip can be divided into multiple functional areas (such as core area, I / O area, etc.) according to the physical location of the chip. Different areas have different force sensitivities, which will affect the subsequent dust removal force adjustment.

[0061] The size of the stain determines the amount of dust removal force that needs to be applied. Larger stains require stronger cleaning force, while smaller stains can use lighter force (for example, the upper force limit of the core area should be smaller than that of the I / O area). The color depth of the stain reflects the stubbornness of the stain. Darker stains may be more difficult to remove and require more cleaning times or greater force.

[0062] Optionally, the system will generate a stain record table, which includes information such as the stain coordinates, the area it belongs to, the size of the area, and the depth of color. This record table will be used for subsequent dust removal operations and quality control.

[0063] When the chip stains are detected, the robot needs to be controlled to go to the dust removal rod placement area to prepare to clamp the dust removal rod. The robot can move between the dust removal rod placement area and the detection area of ​​the dust removal detection machine.

[0064] Among them, the visual system can be used to assist the manipulator in gripping the dust removal rod to ensure the accuracy of gripping, such as controlling the manipulator's gripper to move accurately above the dust removal rod, then descend and clamp the dust removal rod. After gripping the dust removal rod, the manipulator needs to move to a predetermined standby position to prepare for dust removal operations. The standby position is usually a safe position to ensure that the manipulator does not interfere with the chip or other equipment before dust removal.

[0065] Step S50 prepares for the subsequent dust removal operation by recording the stain coordinates of the chip where the stain is detected and controlling the robot.

[0066] As described in step S60, after completing the stain detection of all chips, the system needs to collect the coordinate information of all stains and generate a stain list.

[0067] Each item in the stain list includes information such as the stain's coordinates, chip region, area size, and color depth. At the same time, the standby position coordinates of the manipulator are recorded as the starting point for the shortest path planning.

[0068] Optionally, a shortest path planning algorithm (such as Dijkstra algorithm or A* algorithm) is used to generate the optimal dust removal trajectory. The input includes the coordinates of the standby position and all the stain coordinates, and the output is a shortest path starting from the standby position, passing through all the stain coordinates in sequence, and returning to the standby position or the end point.

[0069] According to the path generated by the shortest path planning algorithm, a detailed dust removal trajectory is generated. The dust removal trajectory includes the order of each stain coordinate, the movement path of the robot, and the cleaning order of each stain.

[0070] After generating the dust removal trajectory, the number of times the dust removal rods need to be replaced can be calculated based on the total area of ​​stains to be cleaned and the upper limit of the cleaning area of ​​a single dust removal rod, and the number of times the robot needs to return to the dust removal rod placement area can be obtained. On this basis, combined with the position of the dust removal rod placement area, the dust removal trajectory can be updated and adjusted.

[0071] Among them, according to the specifications and usage of the dust removal rods, determine the maximum area that each dust removal rod can clean. Calculate the number of dust removal rods that need to be replaced: The number of replacements required = ⌈Total area to be cleaned / Upper limit of cleaning area of ​​a single dust removal rod ⌉; Here, ⌈x⌉ means rounding up.

[0072] This means the number of times the robot returns to the dust removal rod placement area.

[0073] According to the coordinates of the stains, all stains are sorted by position. When planning the dust removal trajectory, the number of times and specific locations of the robot's return to the dust removal rod placement area are determined based on the number of times the dust removal rod needs to be replaced.

[0074] That is, in the dust removal trajectory, insert the path segment of the robot returning to the dust removal rod placement area. For example, if the dust removal rod needs to be replaced once, select a suitable middle position in the dust removal trajectory and insert a path segment returning to the dust removal rod placement area. Combined with the position of the dust removal rod placement area, update and adjust the dust removal trajectory, that is, combine the path segment of the return midway to replan the shortest path of the robot to ensure the shortest total moving distance. Adjust the dust removal trajectory according to the replanned path to ensure that each step is optimized.

[0075] Assumptions: The total area of ​​the stains to be cleaned is 100 square millimeters, and the upper limit of the cleaning area of ​​a single dust removal rod is 40 square millimeters; the position of the dust removal rod placement area is (0, 0); the coordinate positions of all stains are (10, 10), (20, 20), (30, 30), (40, 40), (50, 50).

[0076] Number of replacements required = ⌈100 / 40⌉ = 3.

[0077] Assuming that each time the dust removal rod is replaced, 40 square millimeters can be cleaned, then it is necessary to return to the dust removal rod placement area after cleaning 40 square millimeters and 80 square millimeters. According to the stain distribution, the total stain area of ​​the first and second positions does not exceed 40 square millimeters, and the total stain area of ​​the first, second and third positions exceeds 40 square millimeters. In this case, it is possible to return once after cleaning positions (10, 10) and (20,20); the stain area of ​​the third position does not exceed 40 square millimeters, the total stain area of ​​the third and fourth positions exceeds 40 square millimeters, and the total stain area of ​​the fourth and fifth positions does not exceed 40 square millimeters. In this case, it is possible to return once after cleaning position (30, 30).

[0078] Among them, the shortest path algorithm (such as Dijkstra algorithm) is used to recalculate the total moving path of the robot to ensure the shortest total moving distance. This can effectively optimize the dust removal trajectory of the robot, reduce invalid movement, and improve dust removal efficiency.

[0079] As described in step S70, the robot is controlled to follow the generated dust removal track, and the dust removal rod is used to remove chip stains one by one, and the pressing force of the robot is dynamically adjusted according to the upper limit of the force of each stain and the stain information.

[0080] According to the dust removal track generated in step S60, the control system moves the robot arm along a predetermined path to the coordinate position of each stain in sequence.

[0081] After the robot reaches each stain coordinate, the control system will control the robot to use the dust removal rod to remove the stains in the corresponding chip area.

[0082] Each chip area has an upper force limit, which indicates the maximum pressure that the area can withstand. Exceeding this pressure may cause damage to the chip.

[0083] Optionally, the system needs to adjust the pressing force of the robot according to the size and color depth of the stain. For example, a darker stubborn stain may require a greater pressing force, while a lighter stain may use a smaller pressing force.

[0084] Optionally, a proportional coefficient of the pressing force is set according to the stain area and the color depth. For example, the larger the stain area, the greater the pressing force; the darker the color, the greater the pressing force.

[0085] The upper limit constraint is set to ensure that the calculated pressing force does not exceed the upper force limit of the chip area. The algorithm example is as follows: Pressing pressure = base pressure + (area coefficient × stain area) + (color coefficient × color depth).

[0086] Among them, the base pressure is a default value, and the area coefficient and color coefficient are parameters determined by experiments.

[0087] Optionally, if the calculated pressing force is less than or equal to the upper limit of the force, the calculated pressing force is directly used; if the calculated pressing force is greater than the upper limit of the force, a value less than the upper limit of the force is set as the pressing force parameter.

[0088] The control system adjusts the pressing force of the manipulator in real time according to the upper force limit of each stain and the stain information. A force feedback system or pressure sensor can be used to ensure that the manipulator presses moderately when erasing stains, so as not to damage the chip and effectively remove the stains.

[0089] Optionally, after using the dust removal rod for many times, the system can automatically detect the cleanliness of the dust removal rod and automatically replace it with a new one when necessary to ensure the dust removal effect. Among them, a visual system can be used to assist in detecting the cleanliness of the dust removal rod and automatically replace the dust removal rod through a robot.

[0090] During the dust removal process, the system can monitor the pressing force of the robot and the dust removal effect in real time. If it is detected that the pressing force exceeds the upper limit or the dust removal effect is not ideal, the system can adjust the pressing force of the robot in real time.

[0091] In one embodiment, by combining image recognition, automated mechanical operation and shortest path planning, efficient and accurate automated dust removal of batch chips is achieved, which not only improves the cleanliness of the chips and reduces the generation of defective products, but also improves production efficiency and reduces production costs. This has important application value for the semiconductor manufacturing industry, especially for products such as camera modules that have extremely high cleanliness requirements.

[0092] In one embodiment, based on the above embodiment, the automated dust removal method for batch chips further includes: When the stain area is greater than or equal to the preset area, the robot controls the dust removal rod to perform a single wipe of the stain, so that the dust removal rod contacts the chip surface with a first pressing force, and then gradually increases the pressing force, and rolls the dust removal rod toward the center of the stain until the dust removal rod reaches the center of the stain with a second pressing force; wherein the second pressing force is the upper limit of the force on the chip area; While controlling the dust removal rod to continue rolling forward from the center point of the stain, the pressing force is gradually reduced until the dust removal rod leaves the stain area with the first pressing force; In the process of controlling the rolling of the dust removal rod, the rolling speed is adjusted within a preset speed range according to the color depth of the stain.

[0093] In this embodiment, a more sophisticated dust removal strategy is introduced for larger stains (area greater than or equal to the preset area), that is, by adjusting the pressing force and rolling speed of the dust removal rod, large-area stains can be removed more effectively while protecting the chip from damage caused by excessive pressure.

[0094] For the detected stain, if its area is greater than or equal to the preset threshold, the system will recognize that it is a stain of a larger area and requires special processing methods.

[0095] The robot will accurately position the dust removal rod to the edge of the stain based on the recorded chip location of the stain.

[0096] The dust removal rod first contacts the chip surface with a first pressing force, and this pressure value is pre-set to ensure that no damage is caused to the chip. The first pressing force can be a pre-set or calculated pressing force that is less than the upper limit of the force.

[0097] After the dust removal stick touches the surface, it will gradually increase the pressing force while rolling towards the center of the stain. When the dust removal stick reaches the center of the stain, the pressing force reaches the second pressing force, which is the upper limit of the force in the chip area to ensure that it does not exceed the chip's bearing capacity. This gradual pressurization method helps to gradually and deeply remove the stain and avoid applying too much pressure at one time.

[0098] After that, the dust removal rod continues to roll forward from the center point while gradually reducing the pressing force until it leaves the stained area with the first pressing force. This ensures that no unnecessary pressure is applied to the chip surface after cleaning is completed.

[0099] As the dust removal rod rolls, its speed will be adjusted within a preset speed range. The system can also adjust the rolling speed accordingly based on the color of the stain, such as slowing down for darker stains to ensure more thorough cleaning, and speeding up for lighter stains to improve cleaning efficiency.

[0100] Optionally, within a preset speed range, the higher the color depth, the slower the speed. A speed range is pre-set to allow the dust removal rod to roll and clean within a certain speed range.

[0101] By gradually increasing pressure and then reducing pressure from the edge to the center, a large area of ​​stains can be effectively removed to ensure that the stains are completely removed. At the same time, while controlling the dust removal rod to gradually leave the stained area, the gradual reduction of pressure can avoid vibration or impact caused by the dust removal rod losing contact with the chip surface, which can effectively reduce the mechanical stress on the chip surface and protect the chip surface from damage.

[0102] At the same time, by adjusting the rolling speed according to the color depth of the stains, you can optimize the cleaning for different degrees of pollution and improve the cleaning effect.

[0103] In one embodiment, based on the above embodiment, the automated dust removal method for batch chips further includes: When the stain area is smaller than the preset area, the robot controls the dust removal rod to wipe the stain area at a preset speed; The pressing force used during flat wiping is less than the upper limit of the force in the chip area; and the predetermined speed is set according to the depth of the color of the stain.

[0104] In this embodiment, when the stain area is smaller than the preset area, the robot controls the dust removal rod to perform flat wiping cleaning. Flat wiping means that the dust removal rod moves horizontally in the stain area with a relatively low pressing force, rather than rolling. The pressing force used during flat wiping should be less than the upper limit of the force in the chip area to ensure that the chip is not damaged.

[0105] Optionally, the pressure may be adjusted within a preset range according to the color depth of the stain. For example, for a darker stain, the pressure may be appropriately increased but not exceeding the upper limit; for a lighter stain, the pressure may be appropriately reduced.

[0106] A speed range is preset, and the speed of the dust removal stick when wiping the stained area should be adjusted according to the color depth of the stain. The higher the color depth, the slower the speed: for darker stains, the wiping speed of the dust removal stick should be slower to ensure that the stain is completely removed; for lighter stains, the wiping speed of the dust removal stick can be faster to improve cleaning efficiency.

[0107] For example, for darker stains, the pressing pressure can be set to 90% of the upper limit; for lighter stains, the pressing pressure can be set to 70% of the upper limit; for darker stains, the wiping speed can be set to 10 mm / s; for lighter stains, the wiping speed can be set to 30 mm / s.

[0108] For small stains, flat wiping can complete the cleaning task more quickly and efficiently. Adjust the pressure and speed according to the color depth of the stain to ensure maximum cleaning effect.

[0109] By using a flat-wiping cleaning method when dealing with small-area stains and dynamically adjusting the pressure and speed according to the color depth of the stains, the cleaning effect can be significantly improved while protecting the chip surface from damage. This refined control method is particularly suitable for high-precision manufacturing environments with extremely high cleanliness requirements.

[0110] In one embodiment, based on the above embodiment, after the step of controlling the manipulator to remove chip stains one by one along the dust removal track using the dust removal rod, the method further includes: After the stain erasure of all stain coordinates is completed, the motion camera is controlled to take a re-inspection image for each stain coordinate along the dust removal track; Detect whether the chip stains have been successfully erased based on the re-inspection image; If yes, the corresponding chip is judged as qualified; If not, the robot arm is controlled to use the dust removal rod to perform the stain erasing operation on the corresponding chip again.

[0111] In this embodiment, in order to ensure that the stains on each chip are completely removed, a recheck mechanism is introduced. After the robot completes the erasing operation of all stain coordinates, the motion camera is used to recheck and photograph each stain coordinate position to check whether the stain has been successfully removed. If stains are still found, the robot will be controlled to perform the erasing operation again.

[0112] After the stain erasure of all stain coordinates is completed, the motion camera is controlled to move along the dust removal track to take a re-inspection image for each stain coordinate. After the robot completes the erasure of all stains, the motion camera moves along the same dust removal track to take a picture of each stain coordinate position.

[0113] At each stain coordinate position, a re-inspection image is taken to check whether the stain has been successfully removed. Based on the re-inspection image, image recognition technology is used to detect whether the chip stain has been successfully erased.

[0114] Optionally, if the re-inspected image shows that the stain has been successfully erased, the corresponding chip is determined to be a qualified product.

[0115] Optionally, if the re-inspected image shows that the stain has not been successfully erased, the robot is controlled to use a dust removal rod to perform a stain erasing operation on the corresponding chip again, such as controlling the robot to move to a position where the stain has not been removed again and using the dust removal rod to perform an additional erasing operation.

[0116] Optionally, for stains that are difficult to remove, multiple cleaning attempts can be set, and re-inspection is performed after each cleaning until the stain is removed or the maximum number of cleaning attempts is reached; if the stain is still not removed after multiple cleaning attempts, the system can mark the chip as defective and handle it separately.

[0117] After the robot has finished erasing all the stains, the motion camera is used to recheck and take pictures, and the rechecked images are used to determine whether the stains have been successfully removed. This can effectively improve the cleaning quality and ensure that each chip meets the cleaning standards. For stains that are not completely removed, the robot is controlled to erase again, achieving closed-loop management of the cleaning process and improving production efficiency and product quality.

[0118] In one embodiment, based on the above embodiment, the step of controlling the manipulator to use the dust removal rod to perform the stain erasing operation again on the corresponding chip includes: The contamination coordinates associated with the chip determined as a qualified product are removed from the dust removal track, and the dust removal track is updated; Based on the updated dust removal trajectory, the robot is controlled to use the dust removal rod to uniformly erase the stains on the chips that have not been successfully erased.

[0119] In this embodiment, during the re-inspection process, if it is found that the dirt on some chips has not been successfully wiped off, the system will update the dust removal track and only perform the erasing operation again on the chips that have not been cleared of dirt. This can improve the cleaning efficiency and reduce unnecessary operations.

[0120] After the re-inspection is completed, the system will delete the coordinates of all the stains that are judged as qualified products from the original dust removal track. Then, based on the remaining uncleaned stain coordinates, a new dust removal track is regenerated to ensure that the robot only operates on chips with uncleaned stains, avoiding repeated cleaning of qualified chips and improving efficiency.

[0121] Based on the updated dust removal track, the robot is controlled to use the dust removal rod to uniformly remove the stains on the chips that have not been successfully wiped off. That is, the robot moves to the position of the chips that have not been cleaned of stains according to the updated dust removal track, and uses the dust removal rod to uniformly clean each chip that has not been cleaned of stains.

[0122] Among them, according to the force upper limit of each unremoved stain and the stain information (area size and color depth), the pressing force of the robot is dynamically adjusted to ensure the cleaning effect. It ensures that the unremoved stains can be completely removed, while reducing unnecessary operations and improving cleaning efficiency.

[0123] By updating the dust removal track, the system can centrally clean chips that have not been cleaned of stains, avoiding repeated operations and improving cleaning efficiency. At the same time, through multiple cleaning and re-inspection, it ensures that all stains are completely removed to improve product quality. This combination of dynamic path planning, precise cleaning and automated closed-loop management is particularly suitable for high-precision manufacturing fields with extremely high cleanliness requirements.

[0124] In one embodiment, based on the above embodiment, after the step of detecting whether the chip stain is successfully erased based on the re-inspection image, the method further includes: When it is detected that there is a chip whose stains have not been successfully erased, it is detected whether the dust removal rounds have reached the preset rounds; If the dust removal rounds do not reach the preset rounds, the step of controlling the manipulator to use the dust removal rod to perform the stain erasing operation on the corresponding chip again is executed; If the dust removal rounds have reached the preset rounds, the chips whose stains have not changed before and after erasure are determined as pending products; the chips whose stains have changed before and after erasure are determined as unqualified products; and the automated dust removal operation of the current batch of chips is terminated.

[0125] In this embodiment, the system records the dust removal rounds for each chip. After each cleaning operation, the round counter is incremented by one. When a chip with unremoved stains is detected, the system checks whether the current dust removal round has reached a preset round.

[0126] If the dust removal rounds do not reach the preset rounds, the step of controlling the manipulator to use the dust removal rod to perform the stain erasing operation on the corresponding chip again is executed.

[0127] If the dust removal rounds have reached the preset rounds, the chips whose stains have not changed before and after erasure are determined as pending products; the chips whose stains have changed before and after erasure are determined as unqualified products; and the automated dust removal operation of the current batch of chips is terminated.

[0128] Among them, for chips whose stains have not changed after multiple cleanings, the system will judge them as pending confirmation products. Such chips may require manual inspection or further processing; for stains that have changed but not been completely removed after multiple cleanings, the system will judge them as unqualified products.

[0129] In this way, by setting a preset number of rounds, the system is prevented from trying to clean indefinitely, thus avoiding waste of resources and mechanical wear. The round setting can be adjusted according to actual production conditions to flexibly deal with different types of stains.

[0130] By detecting the changes in the stains before and after erasing, the chips can be accurately classified to avoid misjudgment. For chips that are difficult to judge, they are marked as pending items for manual inspection or subsequent processing.

[0131] In one embodiment, by detecting whether the dust removal rounds have reached the preset rounds, the system can dynamically adjust the cleaning strategy, reduce unnecessary operations, and improve cleaning efficiency. For chips that have reached the preset rounds, the system will classify and process them according to the changes in the stains.

[0132] In one embodiment, based on the above embodiment, the automated dust removal method for batch chips further includes: Generate dust removal results for each chip and display the dust removal results on relevant display devices; sort qualified products, unqualified products and products to be confirmed, and transport qualified products to downstream equipment of the dust removal inspection machine or store them in material boxes, transport unqualified products to the manual dust removal area, and transport unqualified products to the manual inspection area.

[0133] In this embodiment, the system will record the number of cleanings, stain information (position, area, color, etc.) and re-inspection results for each chip. Based on the cleaning history, a dust removal result report for each chip is generated, including the following: chip number or unique identifier, stain removal status (cleared, not cleared, partially cleared), and final judgment result (qualified, unqualified, pending confirmation).

[0134] It should be understood that chips that are not found to be stained during the initial image inspection can be directly determined to be qualified products.

[0135] Optionally, the dust removal results are displayed in real time on a display screen, industrial computer or mobile device on the production line. The display content may include: cleaning statistics of the current batch of chips (qualified rate, unqualified rate, pending confirmation rate), detailed cleaning results of each chip (number, number of cleanings, judgment results), and prompt information of abnormal chips (such as the number and location of chips that have not been cleaned of stains).

[0136] Optionally, a manipulator, conveyor belt or sorting robot is used to classify the chips according to the judgment results. Among them, the chips judged as qualified products are directly transported to the downstream equipment or stored in the material box; the chips judged as unqualified products are transported to the manual dust removal area for further processing; the chips judged as unconfirmed products are transported to the manual inspection area for manual inspection.

[0137] By generating dust removal results and displaying them on relevant equipment, as well as sorting and conveying qualified products, unqualified products and products to be confirmed, the system achieves high efficiency, transparency and traceability of the production process. This solution combines automated cleaning, precise classification and real-time monitoring, and is particularly suitable for high-precision manufacturing fields that require extremely high cleanliness and production efficiency.

[0138] 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 2 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, an automated dust removal method for batch chips as described in the above embodiment is implemented.

[0139] Those skilled in the art will understand that Figure 2 The 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.

[0140] In addition, a dust removal detection machine is also provided in an embodiment of the present application, and the dust removal detection machine includes a control device, a motion camera and a manipulator, and the motion camera and the manipulator both establish a communication connection with the control device and are controlled by the control device; the specific structure of the control device refers to the above embodiment. Since the dust removal detection machine adopts all the technical solutions of all the above embodiments, it has at least all the technical effects brought by the technical solutions of the above embodiments, which will not be repeated here one by one.

[0141] 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 automated dust removal method for batch chips 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.

[0142] In summary, the automated dust removal method, control device, dust removal detection machine and computer-readable storage medium for batch chips provided in the embodiments of the present application realize efficient and accurate automated dust removal for batch chips by combining image recognition, automated mechanical operation and shortest path planning, which not only improves the cleanliness of the chips and reduces the production of defective products, but also improves production efficiency and reduces production costs. This has important application value for the semiconductor manufacturing industry, especially for products such as camera modules that have extremely high cleanliness requirements.

[0143] 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.

[0144] 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.

[0145] 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. An automated dust removal method for batch chips, characterized in that: include: Place the cleaned chips into the detection area of ​​the dust removal detection machine; Controlling the motion camera to take images of the chips in the detection area one by one; wherein the motion camera takes images of each direction of each chip one by one, and after taking images of all directions of the same chip, takes images of the next chip; After all the orientation images of the same chip are captured, all the orientation images of the same chip are synthesized to obtain a chip image of the same chip; Each time a chip image of a chip is obtained, the chip image is subjected to stain detection using image recognition technology; If stains are detected on the chip, the stain coordinates are recorded according to the chip position where the stain is located, and the stain coordinates are associated with the chip area where the stain is located and the stain information, wherein the stain information includes the area size and color depth; and the robot is controlled to go to the dust removal rod placement area to clamp the dust removal rod and move it to the standby position; After completing the stain detection on all chips, the robot's standby position is used as the starting point to connect all the stain coordinates in sequence, and the shortest path planning is performed to generate the dust removal trajectory; The robot is controlled to use a dust removal rod to remove chip stains one by one along a dust removal track; wherein, according to the force upper limit value and stain information of the chip area where each stain is located, the pressing force of the robot when using the dust removal rod to wipe the stain is adjusted.

2. The automated dust removal method for batch chips according to claim 1, characterized in that: The automated dust removal method for batch chips also includes: When the stain area is greater than or equal to the preset area, the robot controls the dust removal rod to perform a single wipe of the stain, so that the dust removal rod contacts the chip surface with a first pressing force, and then gradually increases the pressing force, and rolls the dust removal rod toward the center of the stain until the dust removal rod reaches the center of the stain with a second pressing force; wherein the second pressing force is the upper limit of the force on the chip area; While controlling the dust removal rod to continue rolling forward from the center point of the stain, the pressing force is gradually reduced until the dust removal rod leaves the stain area with the first pressing force; In the process of controlling the rolling of the dust removal rod, the rolling speed is adjusted within a preset speed range according to the color depth of the stain.

3. The automated dust removal method for batch chips according to claim 1, characterized in that: The automated dust removal method for batch chips also includes: When the stain area is smaller than the preset area, the robot controls the dust removal rod to wipe the stain area at a preset speed; The pressing force used during flat wiping is less than the upper limit of the force in the chip area; and the predetermined speed is set according to the depth of the color of the stain.

4. The automated dust removal method for batch chips according to claim 1, characterized in that: After the step of controlling the manipulator to remove chip stains one by one along the dust removal track using the dust removal rod, the method further includes: After the stain erasure of all stain coordinates is completed, the motion camera is controlled to take a re-inspection image for each stain coordinate along the dust removal track; Detect whether the chip stains have been successfully erased based on the re-inspection image; If yes, the corresponding chip is judged as qualified; If not, the robot arm is controlled to use the dust removal rod to perform the stain erasing operation on the corresponding chip again.

5. The automated dust removal method for batch chips according to claim 4, characterized in that: The step of controlling the manipulator to use the dust removal rod to perform a stain erasing operation on the corresponding chip again comprises: The contamination coordinates associated with the chip determined as a qualified product are removed from the dust removal track, and the dust removal track is updated; Based on the updated dust removal trajectory, the robot is controlled to use the dust removal rod to uniformly erase the stains on the chips that have not been successfully erased.

6. The automated dust removal method for batch chips according to claim 4 or 5, characterized in that: After the step of detecting whether the chip stains are successfully erased based on the re-inspection image, the method further includes: When it is detected that there is a chip whose stains have not been successfully erased, it is detected whether the dust removal rounds have reached the preset rounds; If the dust removal rounds do not reach the preset rounds, the step of controlling the manipulator to use the dust removal rod to perform the stain erasing operation on the corresponding chip again is executed; If the dust removal rounds have reached the preset rounds, the chips whose stains have not changed before and after erasure are determined as pending products; the chips whose stains have changed before and after erasure are determined as unqualified products; and the automated dust removal operation of the current batch of chips is terminated.

7. The automated dust removal method for batch chips according to claim 6, characterized in that: The automated dust removal method for batch chips also includes: Generate dust removal results for each chip and display the dust removal results on relevant display devices; sort qualified products, unqualified products and products to be confirmed, and transport qualified products to downstream equipment of the dust removal inspection machine or store them in material boxes, transport unqualified products to the manual dust removal area, and transport unqualified products to the manual inspection area.

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 automated dust removal method for batch chips as described in any one of claims 1 to 7 are implemented.

9. A dust removal detection machine, characterized in that: The dust removal detection machine includes a control device, a motion camera and a manipulator, and the motion camera and the manipulator are both connected to the control device in communication and are controlled by the control device; the control device is the control device as described in 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 automated dust removal method for batch chips according to any one of claims 1 to 7 are implemented.

Images