Positioning method for micro-led array laser rapid repair

Abstract

The application relates to the technical field of Micro LED array laser repair, and discloses a positioning method for Micro LED array laser rapid repair, which comprises the following steps: 1) constructing a coordinate reference system in the imaging field of view of the equipment, taking a fixed angle and a boundary of an array as a reference origin, setting coordinate axes according to horizontal and vertical directions, adjusting the unit length and the coverage range of the coordinate system for Micro LED arrays of different sizes and specifications, and ensuring that the entire array area is covered. The positioning method for Micro LED array laser rapid repair can make the positioning system more matched with the actual physical position of the Micro LED array, reduce the positioning error caused by the inaccuracy of the coordinate system, compare and correct the positioning results of each time of positioning on the defect area, effectively reduce the error accumulation possibly caused by single positioning, and especially when a complex Micro LED array is processed, multiple positioning can continuously optimize the positioning results, so that the finally determined defect position is more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of MicroLED array laser repair, in particular to a positioning method for MicroLED array laser rapid repair. BACKGROUND

[0002] Micro LED array refers to a high-density LED array formed by assembling self-luminous micron-level LEDs as light-emitting pixel units on a driving panel. The main application direction of Micro LED array includes television, computer display, mobile phone screen, tablet computer, wearable device, etc., which can provide high-quality, high-contrast, high-resolution display effect and bring better visual experience to users.

[0003] In the field of modern display technology, MicroLED array is rapidly becoming a strong competitor of the next generation of display technology due to its excellent performance. However, during the manufacturing process of MicroLED array, various defects are inevitably caused due to the complexity of the process and the challenge of the technology. These defects may be caused by multiple links, such as crystal defects in the epitaxial growth process, photolithography deviation in the chip manufacturing process, chip shift or damage during mass transfer, and pollution during packaging, etc. Common defects include non-luminous pixels, non-uniform brightness, color deviation, etc. According to industry statistics, the defect rate in the manufacturing process of MicroLED array is still at a high level, which not only seriously affects the yield of products, but also leads to a substantial increase in production cost, limiting the large-scale commercial application of MicroLED technology. Laser repair technology, as an effective defect repair method, provides a feasible solution to the defect problem of MicroLED array. By accurately controlling the energy and action position of laser, defective MicroLED chips can be repaired, such as repairing open circuit, short circuit and other electrical defects, adjusting brightness and color and other optical properties. However, the key prerequisite for laser repair is to quickly and accurately locate the defect position in the MicroLED array.

[0004] While existing technologies can acquire image information from MicroLED arrays, they struggle to accurately pinpoint defect locations when dealing with high-density, tiny-sized MicroLED pixels due to limitations in resolution and precision, particularly for sub-pixel level defects. Electrical detection-based methods, while capable of detecting pixels with electrical anomalies, suffer from slow detection speeds, hindering rapid repair. Furthermore, interference and crosstalk in complex array structures can affect accuracy. The structural and manufacturing characteristics of MicroLED arrays further complicate defect localization; their tiny pixel sizes and extremely narrow spacing, coupled with variations in structure and parameters between different batches and specifications, place higher demands on the versatility and adaptability of localization methods.

[0005] Therefore, a positioning method for rapid laser repair of MicroLED arrays is needed to solve the problem mentioned in the background that existing positioning methods are unable to accurately determine the specific location of defects. Summary of the Invention

[0006] The purpose of this invention is to provide a positioning method for rapid laser repair of MicroLED arrays, in order to solve the problems mentioned in the background art.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a positioning method for rapid laser repair of MicroLED arrays, comprising: 1) constructing a coordinate reference system within the imaging field of view of the device, taking a fixed angle and boundary of the array as the reference origin, setting coordinate axes in the horizontal and vertical directions, and adjusting the unit length and coverage of the coordinate system for MicroLED arrays of different sizes and specifications to ensure coverage of the entire array area;

[0008] 2) Using high-resolution optical imaging equipment, the MicroLED array is fully scanned and imaged. Based on the preset image analysis algorithm, the system identifies potential defect areas with pixel dimming, abnormal brightness, color deviation and other problems, and performs preliminary defect identification and marking.

[0009] 3) For the initially marked defect areas, a sub-pixel level positioning algorithm is used for precise positioning;

[0010] 4) Multimodal data fusion localization: Electrical detection data is introduced, and machine learning algorithms are used to process the multimodal data to more accurately determine the true location of the defect;

[0011] 5) Coordinate transformation and calibration operation is carried out, a conversion model between the coordinate systems of the imaging device and the laser processing device is established according to the geometric relationship between the two, the imaging device image coordinate system is converted into the motion coordinate system of the laser processing device, and it is ensured that the laser beam accurately irradiates the defect position.

[0012] Preferably, the reference origin in the 1) selects the physical mark easily identified in the Micro LED array, including the process preset cross mark, the specific electrode structure, the packaging boundary corner point or other, and the optical feature extraction includes: collecting the array image through the high-resolution industrial camera and automatically identifying the reference point coordinates using the image processing algorithm.

[0013] Preferably, laser ranging auxiliary positioning can be used between the 2) and the 3), and the height of the Micro LED surface is measured using the confocal laser sensor and the white light interferometer to compensate for the substrate warping or assembly tilt.

[0014] Preferably, the 3) performs sub-pixel interpolation processing on the collected array image, such as bicubic interpolation, and the positioning accuracy is improved to the 0.1 pixel level.

[0015] Preferably, the 4) can also introduce optical image data and other data, and the positions of the electrical performance abnormalities such as open circuit and short circuit can be determined through electrical detection in the electrical detection data, and the electrical abnormal positions are compared and fused with the defect positions identified in the optical image.

[0016] Preferably, the 5) needs to calibrate the laser device and the coordinate system, and the laser device mechanical calibration includes:

[0017] S1: The precision of the motion platform X, Y and Z axes of the device is verified by using a laser interferometer or a grating ruler, and it is ensured that the stepping error is less than ±1 μm;

[0018] S2: The optical path offset of the device is corrected, including aligning the laser focal point with the center of the field of view of the vision system;

[0019] Multi-coordinate system includes:

[0020] S1: The device coordinate system takes the origin of the laser galvanometer or the motion platform as the reference;

[0021] S2: The vision coordinate system is the mapping relationship between the camera imaging plane and the pixel coordinates;

[0022] S3: The workpiece coordinate system is the actual physical coordinates of the Micro LED array, and the multi-coordinate system is realized through the coordinate transformation matrix such as affine transformation or perspective transformation.

[0023] Compared with the prior art, the present application has the following beneficial effects:

[0024] First, the present application can make the positioning system more matched with the actual physical position of the Micro LED array by accurately adjusting the reference coordinate system, reduce the positioning error caused by inaccurate coordinate system, and multiple positioning of the defect area can compare and correct the result of each positioning, effectively reduce the error accumulation that may be generated by single positioning, especially in processing complex Micro LED array, multiple positioning can continuously optimize the positioning result, make the final determined defect position more accurate, multi-modal data fusion positioning combines the data obtained by different types of sensors or detection methods, such as optical image data, electrical detection data, laser scanning data, etc., each kind of data contains different aspects of information about Micro LED array, fusion of these data can accurately judge the defect position from multiple dimensions, make up for the deficiency of single data source in positioning accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 The flowchart of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0027] Please refer to Figure 1 A positioning method for Micro LED array laser rapid repair, 1) a coordinate reference system is constructed in the imaging field of view of the device, a fixed corner of the array is taken as the origin of the reference point, the reference point is preferentially selected at a position with high mechanical stability (such as the edge of the substrate or a fixed pad), to avoid the reference from shifting due to thermal expansion or vibration, the coordinate axes are set according to the horizontal and vertical directions, just like determining the origin and direction on a map, to provide a basic reference for subsequent positioning, for Micro LED arrays of different sizes and specifications, adjust the unit length and coverage range of the coordinate system to ensure that the entire array area is covered, for example, a small size array may use micrometer level unit length, and a large size array may use millimeter level unit length;

[0028] 2) Use high-resolution optical imaging equipment, including high-precision imaging equipment such as high-resolution optical microscopes and CCD cameras, and micro-optical systems (such as long-working-distance objectives) to ensure that the micro details of the Micro LED array can be captured (pixel resolution needs to reach sub-micron level), and the Micro LED array can be fully scanned and imaged. The system identifies potential defect areas with pixel non-light, abnormal brightness, color deviation, and other problems based on pre-set image analysis algorithms, and performs preliminary identification and marking of defects. Once a potential defect location is found, the system will automatically mark it on the image and record the approximate location information of these areas in the image for subsequent more accurate positioning analysis, such as marking the suspected defective pixel points or areas with a specific color frame or point on the image, and enhancing the contrast of defects through specific wavelength illumination (such as blue light or ultraviolet light), for example, dark spots (Dead Pixel) do not emit light when powered on, bright spots (Bright Pixel) continuously emit light, and short-circuit areas may generate infrared signals due to leakage current heating;

[0029] 3) For the preliminary marked defect areas, image preprocessing can be performed first, such as noise reduction (such as Gaussian filtering), contrast enhancement, and background homogenization. Single Micro LED units can be located through edge detection (such as Canny algorithm) or threshold segmentation (Otsu algorithm). Machine learning models (such as convolutional neural networks CNN) can be used to distinguish defect types (dark spots, bright spots, color deviation, etc.) and mark coordinates. Then the results are output to generate a defect coordinate list. Sub-pixel level positioning algorithms can be used for accurate positioning, such as edge detection-based sub-pixel positioning algorithms that use the gray scale variation characteristics of defect edges in images to improve positioning accuracy to sub-pixel level, or template matching-based sub-pixel positioning methods that first create a standard Micro LED pixel template image and then match it in the image containing defects. Based on the similarity and position deviation of the matching, the precise position of the defect is calculated. These algorithms can improve the positioning accuracy from pixel level to sub-pixel level, greatly improving the accuracy of the repair work and enabling more accurate targeting of defect locations.

[0030] 4) Multi-modal data fusion positioning, introducing electrical detection data, using machine learning algorithms to process multi-modal data to more accurately determine the true location of defects. Electrical detection can determine the location of open circuits, short circuits, and other electrical performance abnormalities. These electrical abnormality locations are compared and analyzed with the defect locations identified in the optical image. Machine learning algorithms are used to process multi-modal data to take advantage of different data and more accurately determine the true location of defects, effectively reducing positioning errors caused by the limitations of single detection methods.

[0031] 5) Coordinate transformation and calibration operation is performed, a conversion model between the coordinate systems of the imaging device and the laser processing device is established according to the geometric relationship between the two, the imaging device image coordinate system is converted into the motion coordinate system of the laser processing device, and the laser beam is accurately irradiated to the defect position, a standard calibration plate can be used, which is engraved with feature points of known coordinates, and the calibration plate is measured by the imaging device and the laser processing device to obtain the coordinate data of the two in their respective coordinate systems, so as to accurately calculate the coordinate conversion parameters and ensure the accuracy of the coordinate conversion, providing a reliable position basis for subsequent laser repair. Dynamic positioning adjustment is required during conversion to establish a real-time feedback mechanism.

[0032] In this embodiment, the coordinate system can also be verified and corrected before coordinate conversion and laser device, including selecting multiple test points (such as normally working Micro LED units) in the array, verifying the coordinate accuracy through laser marking or electrical test, and if the deviation exceeds the threshold (such as ±2μm), recalibrating or enabling compensation algorithm.

[0033] In this embodiment, the optimization in the positioning method also includes environmental control and software cooperation, such as operating in a constant temperature (±0.1℃) and shockproof (optical vibration isolation platform) environment to reduce external interference, and realizing automatic calibration process through host computer software (such as LabVIEW or customized Python tool chain).

[0034] In this embodiment, the accurate reference coordinate system and the multiple positioning mechanism help the system to lock the defect area faster, and the system can use the previous positioning information and adjustment experience to find the target more quickly in subsequent positioning, reduce the time cost of search and positioning, and improve the overall repair efficiency. Multi-modal data fusion provides more comprehensive array information, which not only accurately finds the defect position, but also understands the environment and other related information around the defect, which helps to plan a more reasonable laser repair path, avoid unnecessary movement and operation, and further save repair time. Since the positioning is more accurate, the defects can be repaired accurately at one time, reducing repeated repair caused by inaccurate positioning. Repeated repair not only wastes time, but also may cause additional damage to the Micro LED array, so reducing repeated repair can significantly improve repair efficiency.

[0035] Although embodiments of the application have been shown and described, it will be understood by those having ordinary skill in the art that various changes, modifications, alternatives, and variations can be made to these embodiments without departing from the principles and spirit of the application. The scope of the application is defined by the appended claims and their equivalents.

Claims

1. A positioning method for Micro LED array laser rapid repair, characterized in that: Comprise: 1) Construct a coordinate reference system within the imaging field of view of the device, using a fixed angle of the array as the reference origin, and set the coordinate axes in the horizontal and vertical directions. Adjust the unit length and coverage of the coordinate system for Micro LED arrays of different sizes and specifications to ensure coverage of the entire array area; 2) Use high-resolution optical imaging equipment to implement comprehensive scanning imaging of the Micro LED array. The system identifies potential defect areas with pixel non-lighting, abnormal brightness, color deviation, and other problems based on pre-set image analysis algorithms, and performs preliminary identification and marking of defects; 3) For the preliminary marked defect areas, use sub-pixel level positioning algorithms for accurate positioning; 4) Multi-modal data fusion positioning, introduce electrical detection data, use machine learning algorithms to process multi-modal data, more accurately determine the real position of the defect; 5) Coordinate conversion and calibration operation, according to the geometric relationship between the imaging device and the laser processing device, establish the conversion model between the two coordinate systems, convert the imaging device image coordinate system to the motion coordinate system of the laser processing device, ensure that the laser beam accurately irradiates to the defect position; The reference origin in 1) selects physical markers in the Micro LED array that are easy to identify, including process preset cross markers, specific electrode structures, and package boundary corner points. Optical feature extraction includes: collecting array images through high-resolution industrial cameras and automatically identifying reference point coordinates using image processing algorithms; Between 2) and 3), laser ranging assisted positioning can be used. Use confocal laser sensors and white light interferometers to measure the height of the Micro LED surface to compensate for substrate warping or assembly tilt; 4) can also introduce optical image data. In electrical detection data, the positions of electrical performance abnormalities such as open circuits and short circuits can be determined, and these electrical abnormal positions are compared and analyzed with the defect positions identified in the optical image.

2. The positioning method for laser rapid repair of a MicroLED array according to claim 1, characterized in that: 3) performs sub-pixel interpolation processing on the collected array image, improving the positioning accuracy to 0.1 pixel level.

3. The positioning method for laser rapid repair of a MicroLED array according to claim 1, characterized in that: 5) requires calibration of the laser equipment and coordinate system. Laser equipment mechanical calibration includes: S1: Use a laser interferometer or grating ruler to verify the accuracy of the X, Y, and Z axes of the motion platform, ensuring that the step error is less than ±1μm; S2: Correct the optical path offset of the device by aligning the laser focal point with the center of the vision system field of view; Multi-coordinate system alignment includes: S1: Device coordinate system, with the origin of the laser galvanometer or motion platform as the reference; S2: Vision coordinate system, the mapping relationship between the camera imaging plane and the pixel coordinates; S3: Workpiece coordinate system, the actual physical coordinates of the Micro LED array, realize multi-coordinate system alignment through coordinate transformation matrix.

Images