Micro light emitting diode light emitting characteristic detection device and image correction method

Abstract

The application provides a micro light emitting diode light emitting characteristic detection device and an image correction method, and relates to the technical field of semiconductor photoelectric detection. The detection device comprises a synchronous control module, an electrical driving module, an optical acquisition module and an image correction module. The synchronous control module is used to generate an enable signal and a trigger signal; the electrical driving module comprises a sample table for carrying a wafer, a parallel electrical connection interface and a multi-channel programmable power supply, and is used to drive a MicroLED array to emit light simultaneously in response to the enable signal; the optical acquisition module comprises a conic lens and a camera optically coupled with the conic lens, and is used to acquire a light emitting image of the MicroLED array in response to the trigger signal; the conic lens is used to map light rays of different angles to different positions of the light emitting image; and the image correction module is used to perform vignetting compensation and geometric distortion correction on the light emitting image, so as to realize efficient, synchronous and accurate detection of the light emitting characteristics of multiple MicroLEDs.

Description

Technical Field

[0001] This application relates to the field of semiconductor optoelectronic detection technology, specifically to a device for detecting the luminous characteristics of a micro-light-emitting diode and an image correction method. Background Technology

[0002] Microlight-emitting diodes (MicroLEDs), as core components of next-generation display technology, have shown broad application prospects in fields such as augmented reality / virtual reality (AR / VR) near-eye displays, wearable devices, and large-size splicing displays due to their advantages of self-illumination, high brightness, low power consumption, and long lifespan. With the continuous maturation of mass transfer technology and chip manufacturing processes, the size of MicroLED chips has shrunk to below 50μm, allowing millions of chips to be integrated into a single display panel. In the MicroLED manufacturing process, rapid and accurate detection of optical parameters such as the chip's emission angle distribution, luminous intensity, peak brightness, and half-power angle is crucial for ensuring product yield and display uniformity. However, the miniaturization of chip size and the increase in emission angle (typically >120°) pose significant challenges to detection technology: the detection system needs to complete the measurement of the luminous characteristics of thousands of chips within the production line cycle time (typically <10 seconds), while simultaneously capturing the complete light intensity distribution information over a wide angle range.

[0003] In related technologies, single-chip point-by-point rotation measurement and two-dimensional imaging measurement are commonly used to detect the luminous characteristics of MicroLEDs. Specifically, the single-chip point-by-point rotation measurement method involves mounting a light-receiving device on a robotic arm that can rotate around the chip under test. A stepper motor drives the light-receiving device to rotate angularly within a range of 0° to 90°, recording the luminous intensity of the chip at each angular position to construct the luminous angle distribution curve of a single chip. After measurement, a precision displacement platform is used to move to the next chip, repeating the above rotational scanning process until all chips are measured. However, this method suffers from low detection efficiency. Two-dimensional imaging measurement methods employ an imaging colorimeter in conjunction with a macro lens. The camera is perpendicular to the MicroLED wafer surface for single-exposure imaging, simultaneously capturing forward-emitting images of thousands of chips within the field of view. Image processing algorithms extract the luminance and chromaticity coordinates of each chip from a single frame image. However, this method suffers from the loss of large-angle emission information. Furthermore, while using a large-angle conical lens can collect large-angle light, it introduces severe vignetting and edge distortion. The tiny size of MicroLEDs also makes it difficult to achieve synchronous electrical connections and timing control across multiple chips. In other words, existing detection methods struggle to achieve rapid, synchronous, and high-precision optical performance testing of multiple MicroLEDs. Summary of the Invention

[0004] This application provides a device for detecting the luminous characteristics of microLEDs and an image correction method to improve the problem that detection schemes in related technologies are difficult to achieve rapid, synchronous, and high-precision optical performance detection of multiple MicroLEDs.

[0005] In a first aspect, this application provides a device for detecting the luminous characteristics of a micro light-emitting diode, comprising:

[0006] The synchronization control module is used to generate enable and trigger signals;

[0007] The electrical driving module includes a sample stage for supporting MicroLED wafers, a parallel electrical connection interface integrated on the sample stage, and a multi-channel programmable power supply electrically connected to the parallel electrical connection interface. The electrical driving module is electrically connected to the synchronous control module and is used to drive the MicroLED array to emit light simultaneously in response to an enable signal.

[0008] The optical acquisition module includes a conical lens and a camera optically coupled to the conical lens. The optical acquisition module is electrically connected to the synchronization control module and is used to acquire the luminous image of the MicroLED array in response to a trigger signal. The conical lens is used to map light from different angles to different positions of the luminous image.

[0009] The image correction module, electrically connected to the camera, is used to perform vignetting compensation and geometric distortion correction on luminous images.

[0010] In one possible implementation, the conical lens has a viewing angle ≥ ±70°; and / or, the camera has a bit depth ≥ 14 bits and a dynamic range ≥ 70dB.

[0011] In one possible implementation, the electrical drive module further includes a five-axis carrier platform, on which the sample stage is disposed. The five-axis carrier platform includes: an XYZ three-axis translation module for adjusting the position of the MicroLED array in the horizontal and vertical directions; and an θx-θy two-axis tilt adjustment module for adjusting the tilt angle of the MicroLED array relative to the optical axis of the conical lens.

[0012] In one possible implementation, the synchronization control module includes a field-programmable gate array (FPGA), which is used to: output an enable signal to a multi-channel programmable power supply to control the multi-channel programmable power supply to provide drive current during the illumination of the MicroLED array; and output a trigger signal to a camera to control the camera's exposure during the illumination of the MicroLED array; wherein the time interval between the enable signal and the trigger signal is 100μs to 500μs, and the output jitter of the trigger signal is less than 1μs.

[0013] In one possible implementation, the parallel electrical connection interface is a board-to-board (B2B) electrical connection interface; and / or, the sample stage has a vacuum adsorption mechanism for fixing the MicroLED wafer.

[0014] In one possible implementation, the micro-LED luminescence characteristic detection device further includes a filter wheel disposed between the conical lens and the camera to extend the measurement dynamic range.

[0015] Secondly, this application provides an image correction method applied to the image correction module in the aforementioned micro-LED luminescence characteristic detection device. The image correction method includes:

[0016] Acquire images of the emitted light from the MicroLED array;

[0017] Based on the pre-calibrated vignetting coefficient matrix, vignetting compensation is performed on the luminescent image to obtain the vignetting-compensated image;

[0018] Based on the pre-calibrated geometric distortion model, the vignetting compensated image is subjected to geometric distortion correction to obtain the corrected image;

[0019] Extract the emission angle distribution and optical parameters of each MicroLED in the MicroLED array from the corrected image.

[0020] In one possible implementation, the vignetting coefficient matrix is ​​determined as follows: acquiring a vignetting calibration image under uniform light source illumination; extracting the average brightness of the central region of the vignetting calibration image as a reference brightness value; determining the ratio of the brightness value of each pixel in the vignetting calibration image to the reference brightness value to generate an initial vignetting coefficient matrix; and performing smoothing filtering on the initial vignetting coefficient matrix to obtain the vignetting coefficient matrix.

[0021] In one possible implementation, vignetting compensation is performed on the luminous image according to a pre-calibrated vignetting coefficient matrix, including: for each pixel in the luminous image, obtaining the vignetting coefficient corresponding to the pixel's position in the vignetting coefficient matrix, and using the product of the pixel's grayscale value and the vignetting coefficient as the compensated grayscale value corresponding to the pixel to obtain the vignetting-compensated image.

[0022] In one possible implementation, the geometric distortion model is determined as follows: the imaging position of a single MicroLED at different angles is obtained, and a mapping relationship between the actual angle and the radial distance of the image plane is established; the theoretical radial distance corresponding to each angle is determined based on the theoretical mapping relationship when there is no distortion; the deviation between the actual radial distance and the theoretical radial distance is determined to obtain the distortion deviation; and the relationship between the distortion deviation and the theoretical radial distance is fitted using a polynomial to obtain the geometric distortion model.

[0023] In one possible implementation, geometric distortion correction is performed on the vignetting-compensated image according to a pre-calibrated geometric distortion model, including: for each pixel coordinate in the corrected image, determining the normalized coordinates of the pixel coordinates relative to the center of the corrected image, and determining the theoretical radial distance based on the normalized coordinates; obtaining the distortion deviation corresponding to the theoretical radial distance according to the geometric distortion model, and determining the actual radial distance based on the distortion deviation; determining the corresponding source coordinates of the pixel coordinates in the vignetting-compensated image based on the ratio of the actual radial distance to the theoretical radial distance; and sampling the pixel value at the corresponding source coordinates from the vignetting-compensated image as the grayscale value of the pixel coordinates in the corrected image.

[0024] Thirdly, this application provides an image correction device, applied to the image correction module in the aforementioned micro-LED luminescence characteristic detection device, the image correction device comprising:

[0025] The acquisition module is used to acquire the emitted image of the MicroLED array;

[0026] The vignetting compensation module is used to perform vignetting compensation on the luminescent image according to the pre-calibrated vignetting coefficient matrix, so as to obtain the vignetting compensated image;

[0027] The geometric distortion correction module is used to perform geometric distortion correction on the vignetting compensated image according to the pre-calibrated geometric distortion model, so as to obtain the corrected image.

[0028] The parameter extraction module is used to extract the emission angle distribution and optical parameters of each MicroLED in the MicroLED array from the corrected image.

[0029] Fourthly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0030] Memory is used to store instructions executed by the computer;

[0031] A processor for executing computer-executable instructions stored in memory to implement any of the methods of the second aspect.

[0032] Fifthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed, are used to implement the method of the second aspect.

[0033] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed, implements the method of the second aspect.

[0034] In a seventh aspect, this application provides a micro-LED production line-level optical performance testing system, including the micro-LED luminescence characteristic testing device of any of the second aspects above, wherein the single measurement cycle is less than 1 second, and more than 1,500 MicroLEDs can be tested simultaneously in one measurement.

[0035] The microLED luminescence characteristic detection device and image correction method provided in this application include: a synchronization control module for generating an enable signal and a trigger signal; an electrical drive module including a sample stage for supporting the MicroLED wafer, a parallel electrical connection interface integrated on the sample stage, and a multi-channel programmable power supply electrically connected to the parallel electrical connection interface, the electrical drive module being electrically connected to the synchronization control module for simultaneously driving the MicroLED array to emit light in response to the enable signal; an optical acquisition module including a conical lens and a camera optically coupled to the conical lens, the optical acquisition module being electrically connected to the synchronization control module for acquiring the luminescence image of the MicroLED array in response to the trigger signal, the conical lens being used to map light from different angles to different positions of the luminescence image; and an image correction module electrically connected to the camera for performing vignetting compensation and geometric distortion correction on the luminescence image.

[0036] This application uses a synchronous control module to coordinate the electrical drive module and the optical acquisition module in a timely manner, and utilizes a parallel electrical connection interface and a multi-channel programmable power supply to simultaneously drive the MicroLED array to emit light. This enables the parallel detection of multiple MicroLEDs, improving the parallel detection capability of high-density arrays and reducing the problems of long detection cycles and accumulated positioning errors caused by point-by-point scanning, thereby improving detection efficiency. By using a conical lens to map light at different angles to different positions of the emitting image, it is possible to obtain large-angle emitting information of the MicroLED array, thereby improving the ability to characterize the emitting characteristics of the entire field of view. In addition, by using an image correction module to perform vignetting compensation and geometric distortion correction on the emitting image, it is possible to improve the imaging quality and increase the accuracy of measurement results, thereby achieving efficient, synchronous, and accurate detection of the emitting characteristics of multiple MicroLEDs. Attached Figure Description

[0037] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0038] Figure 1 A schematic diagram of a micro-LED luminescence characteristic detection device provided as an exemplary embodiment of this application;

[0039] Figure 2 A schematic diagram of a five-axis vehicle platform provided for an exemplary embodiment of this application;

[0040] Figure 3 Another schematic diagram of the micro-LED luminescence characteristic detection device provided as an exemplary embodiment of this application;

[0041] Figure 4 A flowchart illustrating an image correction method provided as an exemplary embodiment of this application;

[0042] Figure 5 A schematic diagram of the structure of an image correction apparatus provided as an exemplary embodiment of this application;

[0043] Figure 6 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application.

[0044] In the figure, 10—Micro LED luminescence characteristic detection device; 11—Synchronization control module; 12—Electrical drive module; 121—Sample stage; 122—Parallel electrical connection interface; 123—Multi-channel programmable power supply; 124—Five-axis carrier platform; 13—Optical acquisition module; 131—Coniograph lens; 132—Camera; 14—Image correction module; 50—Image correction device; 51—Acquisition module; 52—Vignetting compensation module; 53—Geometric distortion correction module; 54—Parameter extraction module; 60—Electronic equipment; 61—Processor; 62—Memory; 63—Communication interface.

[0045] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0046] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0047] The terms “first,” “second,” etc., used in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, products, or apparatus.

[0048] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0049] In related technologies, existing MicroLED luminescence characteristic detection typically employs single-chip point-by-point rotation measurement and two-dimensional imaging measurement methods. The single-chip point-by-point rotation measurement method, limited by its serial measurement mechanism, requires several hours to detect 1500 chips, resulting in low efficiency and failing to meet the cycle time requirements of mass production line testing. The two-dimensional imaging measurement method, due to the limited numerical aperture of the lens, cannot completely collect luminescence information greater than ±30°, leading to a lack of characterization of MicroLED's large-angle luminescence characteristics. While using a large-angle conical lens can collect large-angle light, it introduces severe vignetting (edge ​​brightness attenuation can reach over 50%) and edge distortion (angle measurement error can reach over 2.5°), affecting photometric measurement accuracy. Furthermore, the tiny size of MicroLED chips (less than 50μm) makes it difficult for traditional probe cards to achieve high-density, parallel electrical connections and microsecond-level timing synchronization control of multiple chips, and there is a risk of contact damage to the chip surface. Therefore, existing detection solutions struggle to achieve rapid, synchronous, and high-precision optical performance detection of multiple MicroLEDs.

[0050] To address the aforementioned issues, this application provides a micro-LED luminescence characteristic detection scheme. A synchronous control module generates enable and trigger signals, which work in conjunction with an electrical drive module and an optical acquisition module. The electrical drive module includes a sample stage for supporting the MicroLED wafer, a parallel electrical connection interface integrated on the sample stage, and a multi-channel programmable power supply electrically connected to the parallel electrical connection interface. This power supply responds to the enable signal and simultaneously drives the MicroLED array to emit light. The optical acquisition module includes a conical lens and a camera optically coupled to the conical lens. This module acquires the emitted light image under the trigger signal and maps light from different angles to different positions in the image. An image correction module is electrically connected to the camera and performs vignetting compensation and geometric distortion correction on the emitted image. This technical approach enables large-angle optical information acquisition while maintaining synchronous emission of the MicroLED array, and allows for necessary image correction. This allows for rapid and accurate detection of multiple MicroLEDs simultaneously, effectively improving the problem of related technologies struggling to achieve rapid, synchronous, and high-precision optical performance detection of multiple MicroLEDs.

[0051] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0052] Figure 1 This is a schematic diagram of a micro-LED luminescence characteristic detection device provided as an exemplary embodiment of this application. Figure 1 As shown, the micro-LED luminescence characteristic detection device 10 includes: a synchronization control module 11, an electrical drive module 12, an optical acquisition module 13, and an image correction module 14, wherein:

[0053] Synchronous control module 11 is used to generate enable signal and trigger signal;

[0054] The electrical drive module 12 includes a sample stage 121 for supporting MicroLED wafers, a parallel electrical connection interface 122 integrated on the sample stage 121, and a multi-channel programmable power supply 123 electrically connected to the parallel electrical connection interface 122. The electrical drive module 12 is electrically connected to the synchronization control module 11 and is used to drive the MicroLED array to emit light simultaneously in response to an enable signal.

[0055] The optical acquisition module 13 includes a conical lens 131 and a camera 132 optically coupled to the conical lens 131. The optical acquisition module 13 is electrically connected to the synchronization control module 11 and is used to acquire the light-emitting image of the MicroLED array in response to a trigger signal. The conical lens 131 is used to map light from different angles to different positions of the light-emitting image.

[0056] Image correction module 14, electrically connected to camera 132, is used to perform vignetting compensation and geometric distortion correction on luminous images.

[0057] Among them, the MicroLED array refers to a collection of multiple MicroLEDs arranged according to a predetermined rule on the same substrate or wafer-level substrate. Its typical characteristics are small single chip size, high array density, fast light emission response and wide angular distribution. The light emission image refers to the two-dimensional digital image data recorded by the optical acquisition module 13 after the electrical driving module 12 applies excitation to the MicroLED array. The image not only contains the spatial position and brightness information of each MicroLED, but also implies the image plane distribution relationship corresponding to different light emission angles due to the angular mapping characteristics of the conical lens 131.

[0058] Correspondingly, the synchronization control module 11 is a timing control unit used to generate enable and trigger signals. Its function is to coordinate the start time, duration window, and stop time of electrical drive and image acquisition, so that the MicroLED array is lit up within the predetermined light-emitting range and accurately exposed by the camera 132. The synchronization control module 11 can be set in the control cabinet, control board, or embedded control box in the micro LED light-emitting characteristic detection device 10, and is electrically connected to the multi-channel programmable power supply 123 and the camera 132 through cables, ribbon cables, buses, or high-speed communication interfaces. In terms of form, it can be a field programmable gate array control board, an independent timing generator module, or an integrated motion control motherboard. In terms of size, it can be set as a single board, rack, or embedded module structure according to the integration space of the production line, and form a plug-in or external arrangement with the camera acquisition card to ensure the stability and anti-interference capability of the timing output.

[0059] The electrical drive module 12 includes a sample stage 121 for supporting MicroLED wafers, a parallel electrical connection interface 122 integrated on the sample stage 121, and a multi-channel programmable power supply 123 electrically connected to the parallel electrical connection interface 122. The electrical drive module 12 is electrically connected to the synchronization control module 11 and is used to simultaneously drive the MicroLED array to emit light in response to an enable signal. The sample stage 121 is used to place the MicroLED wafers or packaging substrates and form a stable support reference. Its position is usually located in the sample support area at the bottom of the device and is correspondingly set with the optical acquisition module 13 above it along the same detection optical axis. The parallel electrical connection interface 122 is used to synchronously introduce multiple driving currents into the array under test. It can be a board-to-board connector array, a flexible pin array, a gold-plated pad interface, or a flexible circuit adapter. The multi-channel programmable power supply 123 can be an independent modular constant current source, an integrated multi-channel driver board, or a rack-mount power supply unit, so as to provide parallel power supply to multiple light-emitting units according to the layout of the array under test. The sample stage 121 can be a flat vacuum sample stage, a ring-shaped support stage, or a tray-type sample holder with a positioning reference surface. The main materials can be aluminum alloy, hard anodized aluminum, stainless steel, or ceramic insulating board. The plane size of the sample stage is usually larger than the outer size of the wafer under test to reserve installation and alignment margin. The pin spacing of the connection interface is matched with the pad pitch of the MicroLED array or flexible circuit board. The number of power supply channels is configured according to the number of chips tested in a single operation, so that each channel outputs drive current synchronously when the enable signal arrives, and shuts off the output in sequence after the light emission is completed.

[0060] The optical acquisition module 13 includes a conical lens 131 and a camera 132 optically coupled to the conical lens 131. The optical acquisition module 13 is electrically connected to the synchronization control module 11 and is used to acquire the emission image of the MicroLED array in response to a trigger signal. The conical lens 131 is used to map light rays at different angles to different positions of the emission image, so as to convert the angle information into imageable spatial distribution information, thereby realizing the characterization of the large-angle emission characteristics. The optical acquisition module 13 is usually arranged above the sample stage or at the detection optical axis. The conical lens 131 is connected to the camera 132 through a C-mount, F-mount, bayonet mount, or custom flange. Then, the camera 132 is connected to the acquisition card or the main control computer via a data cable. The lens structure can be a monotube conical lens, a conical lens assembly with an additional compensation lens group, or a composite lens structure with an integrated relay lens group. The lens body material can be a combination of aluminum alloy lens barrel, optical glass lens, coated lens and blackened light shield, or aspherical glass element, resin lens or microlens array enhancement structure. The light transmission diameter, focal length and camera target surface size satisfy the imaging coverage relationship. The working distance from the front end of the lens to the sample stage 121 is matched with the height of the device to ensure that large-angle light can enter the effective aperture and be mapped to the corresponding position.

[0061] Image correction module 14 is electrically connected to camera 132 and is used to perform vignetting compensation and geometric distortion correction on the luminous image. Its function is to correct brightness and position deviations caused by transmittance attenuation at the edge of the conical lens, imaging nonlinearity, and field-of-view distortion, thereby restoring the correspondence between pixel brightness and actual angle. This module is typically integrated as a software algorithm module into the main control computer, image processing workstation, or graphics processing unit (GPU). The image correction module 14 is located in a GPU acceleration unit and receives the original image output by the camera 132 through a data interface. The carrier can be an industrial control computer hard disk, solid-state storage, field-programmable gate array coprocessor board, or graphics processor computing card. The algorithm parameters can be stored based on calibration data table, lookup table, or polynomial model, and the computing resource configuration is matched with the resolution, frame rate, and scale of the array under test of the camera 132 to ensure that the compensation and correction of a single frame image are completed within a preset processing time limit. The image correction module 14 can also be implemented by image correction based on neural network, geometric correction based on spline function, or multi-level correction based on calibration board. However, in this embodiment, it is preferred to perform vignetting compensation first, then perform geometric distortion inversion mapping, and then output the corrected image data.

[0062] Based on the above analysis, it can be seen that in the operation of the micro LED luminescence characteristic detection device provided in this application embodiment, when the system is started, the synchronous control module 11 first receives the start command from the host computer and generates an enable signal and a trigger signal according to a preset timing sequence. When the enable signal arrives, the electrical drive module 12 synchronously applies a driving current to the MicroLED array on the sample stage 121 through the parallel electrical connection interface 122, so that the large-scale array enters the luminescence state within a unified time window. At the same time, the optical acquisition module 13 starts the camera 132 for exposure under the action of the trigger signal. The conical lens 131 maps the light rays with different emission angles to different spatial positions in the luminescence image according to the angular distribution, so that the large-angle luminescence information that is originally difficult to be completely covered by ordinary imaging lenses can be acquired in the form of a two-dimensional image. Subsequently, the image correction module 14 first performs vignetting compensation on the obtained original image to correct the brightness attenuation caused by the change in transmittance at the edge of the field of view, and then performs geometric distortion correction to eliminate the mapping nonlinearity and position offset introduced by the conical lens, so that the corrected image can more accurately reflect the true luminescence characteristics of the MicroLED array at different angles. Through the above technical means, the device can improve the integrity of large-angle light emission information acquisition while maintaining array synchronous driving and overall acquisition efficiency, and reduce the measurement deviation at the edge of the field of view. This makes the brightness distribution, angle distribution and uniformity analysis results closer to the actual state of the device, and is suitable for online or offline batch testing of MicroLED wafers, mass transfer arrays and packaged devices.

[0063] The micro-LED luminescence characteristic detection device provided in this application uses a synchronous control module to coordinate the electrical drive module and the optical acquisition module in a timely manner, and utilizes a parallel electrical connection interface and a multi-channel programmable power supply to simultaneously drive the MicroLED array to emit light. This enables the parallel detection of multiple MicroLEDs, improving the parallel detection capability of high-density arrays and reducing the problems of long detection cycles and accumulated positioning errors caused by point-by-point scanning, thereby improving detection efficiency. By using a conical lens to map light at different angles to different positions of the luminescence image, it is possible to acquire large-angle luminescence information of the MicroLED array, thereby improving the full-field luminescence characteristic characterization capability. In addition, by using an image correction module to perform vignetting compensation and geometric distortion correction on the luminescence image, it is possible to improve imaging quality and increase the accuracy of measurement results, thereby achieving efficient, synchronous, and accurate detection of the luminescence characteristics of multiple MicroLEDs.

[0064] In some embodiments, the angle of view of the conical lens is ≥ ±70°; and / or, the bit depth of the camera is ≥ 14 bits, and the dynamic range of the camera is ≥ 70dB.

[0065] In one possible embodiment, a conical lens is an optical imaging component used to map light rays from a MicroLED array under test at different emission angles to different positions on the image plane according to angular information. Its function is to expand the range of emission angles covered by a single imaging, allowing light rays from large angle directions to still enter the camera imaging system and retain angular distribution characteristics, thereby providing a raw image basis for subsequent brightness distribution analysis, angle response analysis, and field-of-view edge correction. The conical lens is positioned at the front end of the optical acquisition module, with the front end facing the MicroLED wafer or array carrier on the sample stage. The rear end is correspondingly positioned in the optical axis direction with the photosensitive target surface of the camera and maintains an optical coupling relationship, so that light rays from different incident directions are refracted by a lens group and projected onto the corresponding areas on the camera imaging surface. For example, the viewing angle of the conical lens is ≥±70° (full angle ≥140°), which can simultaneously collect the large-angle light emission signals of MicroLEDs, helping to reduce the truncation of edge angle information under ordinary lens conditions; the working distance of the lens is matched with the thickness of the MicroLED wafer (typically 10mm), ensuring that more than 1500 MicroLEDs can be covered simultaneously in the object-side field of view (calculated based on a 30μm chip size and 50μm pitch, the field of view is ≥5mm×6mm).

[0066] In one possible embodiment, the camera is an area array imaging device optically coupled to the rear end of a conical lens. Its function is to perform short-time exposure and digital acquisition of the emitted image output by the conical lens under the action of a trigger signal, and to record the light intensity changes at different brightness levels with high grayscale resolution, thereby reducing noise overload in low-brightness areas or distortion due to saturation in high-brightness areas. The camera is mounted behind the lens and coaxially fixed to the lens via a mechanical bracket, flange interface, or positioning sleeve to ensure the relative position stability between the lens image plane and the camera target plane, thus ensuring a consistent mapping relationship between angular information and image plane position. The camera's bit depth is greater than or equal to 14 bits, indicating that its single-frame image can output at least approximately 16,000 grayscale discrete levels, thereby improving the grayscale resolution capability for MicroLEDs with different luminous intensities. Simultaneously, its dynamic range is greater than or equal to 70 dB, meaning that the camera has a wide exposure tolerance, capable of simultaneously retaining identifiable information of both darker and brighter chips in the same image. For example, the camera can be a 14-bit cooled scientific-grade complementary metal-oxide-semiconductor (sCMOS) camera with a dynamic range ≥70dB, a pixel resolution ≥4 megapixels (typical value 2048×2048), and a pixel size of 6.5μm. The camera connects to the image acquisition card via Camera Link or CoaXPress interface and supports external trigger mode, with the trigger signal provided by the synchronization control module.

[0067] When the system starts, the synchronization control module first sends an enable signal to the electrical drive module, causing the MicroLED array on the sample stage to light up synchronously under the combined action of the parallel electrical connection interface and the multi-channel programmable power supply. Then, the synchronization control module sends a trigger signal to the optical acquisition module, driving the conical lens at the front end to receive emitted light from different angles of the MicroLED array and map the angular information within a range of at least ±70° onto the corresponding position on the camera's imaging surface. Because the camera has a bit depth of at least 14 bits and a dynamic range of at least 70dB, it can simultaneously retain the weaker light signal at the edge of the field of view and the stronger light signal in the central region during a single exposure, reducing detail loss caused by saturation, insufficient quantization, or noise interference. This allows the image correction module to perform vignetting compensation and geometric distortion correction based on a more complete original image. Therefore, while maintaining the array's synchronous drive and overall acquisition efficiency, it can more fully capture the large-angle emission characteristics of MicroLEDs and improve the grayscale representation capability across the entire field of view, thus facilitating more accurate subsequent analysis of brightness distribution, angular response, and uniformity. Based on the above analysis, it can be seen that this method enables large-scale MicroLED array detection to still have good measurement integrity and repeatability in edge regions and scenarios with large differences in brightness.

[0068] In this embodiment, by employing a conical lens with a viewing angle ≥ ±70°, the large-angle luminous information of the MicroLED array can be completely captured in a single exposure, effectively overcoming the technical deficiency of traditional lenses that cannot collect luminous information greater than ±30°, and significantly improving the characterization capability of the full-field luminous characteristics. In addition, by employing a high dynamic range camera with a bit depth ≥ 14bit and a dynamic range ≥ 70dB, it is possible to simultaneously retain the weak edge light signal and the strong center light signal, which helps to reduce the noise-overwhelmed low-brightness area or the saturation distortion of the high-brightness area, and significantly expands the brightness response range of the measurement system. Through the synergistic effect of the two, a more complete original image basis is provided for subsequent vignetting compensation and geometric distortion correction, thereby improving the detection accuracy and achieving efficient and accurate detection of the luminous characteristics of the MicroLED array.

[0069] In some embodiments, the electrical drive module further includes a five-axis carrier platform, on which the sample stage is disposed. The five-axis carrier platform includes: an XYZ three-axis translation module for adjusting the position of the MicroLED array in the horizontal and vertical directions; and an θx-θy two-axis tilt adjustment module for adjusting the tilt angle of the MicroLED array relative to the optical axis of the conical lens.

[0070] The sample stage is mounted on a five-axis carrier platform, which includes an XYZ three-axis translation module and an θx-θy two-axis tilt adjustment module. The XYZ translation module is used for linear displacement adjustment. The X and Y axes are used for lateral positioning of the sample stage in the horizontal plane, while the Z axis is used for focal length compensation and working distance adjustment in the vertical direction, ensuring that the MicroLED array under test is within the effective imaging area of ​​the conical lens. The θx-θy two-axis tilt adjustment module is used for dual-axis attitude fine-tuning to compensate for wafer warpage, mounting misalignment, or platform errors. This maintains a set relative relationship between the emitting surface of the MicroLED array and the optical axis of the conical lens, thereby reducing field-of-view shift and angle mapping errors caused by tilt. For example, the θx-θy two-axis tilt adjustment module fine-tunes the wafer attitude to maintain an orthogonal or set tilt angle relationship between the emitting surface of the MicroLED array and the optical axis of the conical lens, thus compensating for wafer warpage, mounting misalignment, or platform errors. For example, when the wafer warps locally due to thermal expansion, the θx-θy adjustment module can make minute adjustments to the angle so that the imaging area of ​​the conical lens always covers the effective light-emitting area of ​​the MicroLED array, reducing field of view offset and angle mapping error.

[0071] Figure 2 A schematic diagram of a five-axis vehicle platform provided for an exemplary embodiment of this application. Figure 2 As shown, the five-axis carrier platform is used to support the sample stage and drive its movement. For example, the XYZ three-axis translation module has a travel of 100mm × 100mm × 20mm, a positioning accuracy ≤1μm, and a repeatability ≤0.5μm; the θx-θy two-axis tilt adjustment module has an adjustment range of ±5° and a resolution of 0.001°; the five-axis carrier platform is connected to the main control computer via a motion control card (such as an EtherCAT bus) to achieve closed-loop control.

[0072] Figure 3 Another schematic diagram of the micro-LED luminescence characteristic detection device provided as an exemplary embodiment of this application. (See diagram below.) Figure 3 As shown, the sample stage 121 is mounted on a five-axis carrier platform 124, wherein the sample stage 121 is used to support MicroLED wafers; the five-axis carrier platform 124 is configured to be compatible with a variety of standard wafer sizes, including 4-inch, 6-inch and 8-inch wafers.

[0073] Correspondingly, when the system starts, the MicroLED wafer is first fixed on the sample stage and then mounted on the five-axis carrier platform along with the sample stage. The control system drives the XYZ three-axis translation module to center the wafer, correct its horizontal position, and adjust its height according to preset detection parameters, so that the array under test enters the effective working distance range of the conical lens. Subsequently, the θx-θy two-axis tilt adjustment module performs minor corrections to the wafer's attitude to compensate for wafer warpage, clamping deviation, and platform installation errors, so that the emitting surface is as orthogonal as possible to the lens optical axis or has a set tilt angle relationship. Since the sample stage can achieve continuous fine adjustment in five degrees of freedom, it can ensure that the wafer is centered in the imaging process and reduce edge defocus and angular deviation caused by uneven surface shape. This makes the light emission information collected by the conical lens from different angles more complete, and the brightness attenuation and geometric distortion at the edge of the field of view easier to correct, thereby improving the detection consistency, repeatability, and large field of view measurement accuracy of the MicroLED array's light emission characteristics. Based on the above analysis, it can be seen that the five-axis carrier platform can achieve precise matching of sample position and orientation without significantly increasing the complexity of mechanical scanning, and is suitable for batch optical inspection of MicroLED wafers, mass transfer arrays and packaged devices.

[0074] In some embodiments, the synchronization control module includes a field-programmable gate array (FPGA) for: outputting an enable signal to a multi-channel programmable power supply to control the multi-channel programmable power supply to provide drive current during the illumination of the MicroLED array; and outputting a trigger signal to a camera to control the camera's exposure during the illumination of the MicroLED array; wherein the time interval between the enable signal and the trigger signal is 100 μs to 500 μs, and the output jitter of the trigger signal is less than 1 μs.

[0075] In one possible embodiment, the synchronization control module includes a field-programmable gate array (FPGA) for uniformly arranging the power-on, gating, and camera exposure triggering of the multi-channel programmable power supply during MicroLED light emission detection, ensuring a fixed timing relationship between electrical drive and optical acquisition. The FPGA is connected to the enable terminal of the multi-channel programmable power supply and the external trigger terminal of the camera via digital output interfaces. The interface can be of transistor-transistor logic (TTL) level, low-voltage transistor-transistor logic (LVTTL) level, or an optocoupler-isolated trigger interface. Field-programmable gate arrays (FPGAs) can be implemented using standalone FPGA boards, System-on-Chips (SoC) integrated control modules, or timing control modules with external crystal oscillators and phase-locked loops (PLLs). Their internal logic units generate preset delays, pulse widths, and synchronization phases, ensuring the enable signal precedes the trigger signal, with a time interval of 100μs to 500μs between them. This allows the multi-channel programmable power supply to complete voltage regulation, channel selection, and current ramp-up before the output drive current, enabling the camera to begin exposure after the MicroLED array has entered a stable emission state. For the trigger signal, the FPGA generates a low-jitter pulse through direct hardware logic output, with output jitter less than 1μs. This ensures high-precision alignment between the camera's exposure time and the MicroLED array's emission time. Excessive jitter can cause misalignment between the exposure and emission windows, resulting in image brightness deviations or inter-frame differences. Controlling jitter through hardware synchronization ensures the repeatability of brightness distribution and angle response data in a single measurement, contributing to improved stability in large-scale array detection.

[0076] For example, the time synchronization controller is implemented using a field-programmable gate array (FPGA) such as the Xilinx Artix-7 series. Its input signals include an external start signal and a power-ready signal from the host computer; its output signals include a camera trigger signal (TTL level, rising edge triggered, jitter <1μs) and a power enable signal. The FPGA communicates with the host computer via Universal Serial Bus (USB) 3.0 or Ethernet to receive measurement parameter configurations. Correspondingly, the programmable power supply, for example, uses a multi-channel precision current source (such as the Keithley 2600B series or a self-designed multi-channel constant current source) with 32 independent output channels, each channel outputting a current range of 0.1mA-100mA, a resolution of 16 bits, and a settling time of <10μs. The programmable power supply connects to the host computer via a General Purpose Interface Bus (GPIB) or Local Area Network (LAN) interface to receive current setpoints and switching commands.

[0077] When the system starts, the field-programmable gate array (FPGA) first receives the start command sent by the host computer or detection program, and outputs an enable signal to the multi-channel programmable power supply according to the pre-written timing parameters. This enables the power supply to enter the working state and outputs the driving current of each channel to the parallel electrical connection interface integrated on the sample stage. The MicroLED array emits light synchronously under the action of this driving current. Subsequently, after a set delay of 100μs to 500μs, the FPGA outputs a trigger signal to the camera, enabling the camera to complete the exposure acquisition during the illumination of the MicroLED array. Because the enable signal is output first and sufficient time is allowed for the power supply to complete the output stabilization, the illumination state corresponding to the camera trigger is in a relatively stable working range, which can reduce the impact of transient fluctuations during the power supply setup process on image brightness and consistency. At the same time, because the trigger signal adopts a hardware synchronization method with less than 1μs jitter, the camera exposure start time and the array illumination time can be kept in high precision, reducing brightness deviation and inter-frame differences caused by timing drift. Based on the above action process, it can be seen that this synchronous control method can improve the temporal consistency and repeatability of MicroLED array light emission images while maintaining a short batch detection cycle, and provide time-stable original image data for subsequent vignetting compensation and geometric distortion correction, thereby improving the detection accuracy and repeatability of large-scale array light emission characteristics.

[0078] In some embodiments, the parallel electrical connection interface is a board-to-board (B2B) electrical connection interface; and / or, the sample stage has a vacuum adsorption mechanism for fixing the MicroLED wafer.

[0079] In one possible embodiment, the B2B electrical connection interface is a high-density electrical connection component used for direct interlocking between upper and lower boards or adjacent boards, enabling the establishment of parallel electrical paths between the MicroLED wafer and a multi-channel programmable power supply. The interface's function is to connect the driving channels corresponding to each MicroLED unit in parallel, thereby ensuring electrical connection consistency while reducing indentation, wear, and contact instability problems introduced by point-by-point probe contact. Its mounting location is typically set in the bearing area of ​​the sample stage, the upper surface of the adapter carrier, or at the interface between the sample stage and the back electrode of the wafer. During assembly, it forms a relatively fixed relationship with the sample stage through positioning posts, limiting grooves, or floating compensation structures, and forms an electrical interface with the pads, flexible circuit board, or intermediate adapter board on the back of the MicroLED wafer. The B2B electrical connection interface can adopt a through-hole structure, a floating compensation structure, or an array structure with flexible contact terminals. The terminal materials may include copper alloy conductors, gold-plated contact layers, nickel underlayers, and wear-resistant insulating shells. It can also adopt contact arms with springs, split male and female connector structures, or high-density crimp connection structures to adapt to different array pitches and assembly forms. For example, the pin pitch of the interface can be set to a small pitch that matches the pitch of the MicroLED array pads, typically 0.2mm to 1.0mm. The connection height and crimp stroke are usually controlled within a range that can achieve stable conduction while avoiding overvoltage damage to the wafer electrodes. For example, the crimp stroke can be 0.1mm to 0.5mm. The terminal contact surface can adopt a gold plating layer, a nickel barrier layer, and a copper conductor stack structure to improve conductivity, oxidation resistance, and durability. Based on the above structure, when the sample stage is aligned and pressure is applied, each channel can form a reliable parallel electrical connection with the wafer array through the B2B electrical connection interface, thereby realizing the synchronous driving of multiple MicroLEDs during the light emission test and improving the connection repeatability and test stability.

[0080] The sample stage features a vacuum adsorption mechanism for securing MicroLED wafers. This mechanism is defined as having a negative pressure cavity, an array of adsorption holes, and vacuum tubing inside or on the surface of the sample stage. An external vacuum pump generates negative pressure, creating an adsorption force between the back of the wafer and the stage surface. This suppresses wafer slippage, warping, and micro-vibrations during the wafer's emission and optical acquisition processes, maintaining relative stability between the wafer, the parallel electrical interface, and the optical axis of the conical lens. For example, when the wafer shifts due to mechanical vibration, the vacuum adsorption force maintains relative stability between the wafer and the optical axis of the conical lens, preventing field-of-view shifts and image deviations caused by posture changes. The adsorption hole array corresponds to the effective testing area of ​​the wafer, and the vacuum tubing is connected to the vacuum pump, negative pressure control valve, and control module. For example, the diameter of the adsorption pores can be set from 0.2 mm to 2 mm, and the pore spacing can be arranged from 2 mm to 20 mm according to the wafer size and negative pressure uniformity requirements. The adsorption area preferably covers the entire effective bearing area of ​​the wafer, and the vacuum degree is less than or equal to -80 kPa to ensure stable adsorption without causing local deformation of the wafer. The mechanism can also be further equipped with a zoned control loop to obtain a relatively uniform adsorption force between the wafer edge and the central area, thereby improving the flatness and repeatability of large-size wafers during inspection.

[0081] For example, the sample stage features an electrical interface area in its center for integrating a high-density B2B electrical connection interface (0.4mm pin pitch, ≥1600 pins), which can directly interlock with the flexible printed circuit board (FPCB) of the MicroLED. This interface connects to a multi-channel gating circuit board via the flexible circuit board, which in turn connects to the main control computer via a general purpose input / output (GPIO) interface to achieve row and column gating control. The sample stage also includes a vacuum adsorption mechanism with a vacuum level ≤-80kPa for fixing the MicroLED wafer or FPCB.

[0082] When the system starts, the sample stage first establishes negative pressure adsorption under the action of the vacuum pump, which firmly attaches the MicroLED wafer to the stage surface and restricts its orientation change. Subsequently, the sample stage mechanically aligns the pads or intermediate transfer board on the back of the wafer with the board-to-board B2B electrical connection interface to complete the mating contact, forming a multi-channel parallel electrical path. After the synchronous control module sends an enable signal, the multi-channel programmable power supply synchronously outputs driving current to the MicroLED array through the connection interface. Each light-emitting unit of the wafer simultaneously enters the light-emitting state under controlled timing, while the optical acquisition module collects light from different angles through the conical lens under the action of the trigger signal and forms a light-emitting image. Because the vacuum adsorption mechanism restricts the displacement of the wafer during the light emission and acquisition process, and the B2B electrical connection interface ensures stable contact of the high-density channels, the entire system can complete the synchronous lighting and overall imaging of a large-scale array in a short time. This reduces the mechanical damage and positioning error caused by the point-by-point contact of traditional probes, and reduces the image deviation caused by wafer warping, poor contact, or orientation shift. This is beneficial to improving the detection efficiency, connection reliability, and measurement repeatability of MicroLED light emission characteristics.

[0083] Based on the above embodiments, in some embodiments, the micro-LED luminescence characteristic detection device further includes a filter wheel, which is disposed between the conical lens and the camera to extend the measurement dynamic range.

[0084] In one possible embodiment, a filter wheel is positioned between the conical lens and the camera to adjust the luminous flux entering the camera based on the luminous intensity of the MicroLED array under test. This prevents pixel saturation in high-brightness samples and maintains a sufficient signal-to-noise ratio in low-brightness samples, making the same detection device applicable to test scenarios with different brightness levels and emission bands. The filter wheel is typically coaxially arranged with the conical lens and the camera to ensure that the angular information mapped by the conical lens is only affected by predetermined attenuation or band filtering before entering the camera, without introducing additional field-of-view shift. This wheel can be implemented using a motor-driven disc structure, a tower-type switching mechanism, or a linear insert-type switching mechanism. The disc structure is suitable for arranging more filters and achieving rapid switching, the tower-type switching mechanism is suitable for miniaturized installation, and the linear insert-type switching mechanism facilitates multi-level attenuation switching within a limited space. The filter body can be made of glass substrate with metal oxide coating, resin-based filter, or composite multilayer dielectric film. Among these, glass substrate structure has high optical stability and heat resistance, resin-based filter facilitates lightweight design, and composite multilayer dielectric film structure is suitable for forming multiple transmittance levels and improving spectral selectivity. The light transmission aperture of the filter should generally not be smaller than the effective imaging aperture of the conical lens to avoid additional vignetting caused by light cutoff.

[0085] For example, the filter wheel is equipped with multiple neutral density (ND) filters (such as ND0.3 to ND3.0) to extend the measurement dynamic range. It is driven by a stepper motor and communicates with the control system via RS-232 (Recommended Standard 232). The wheel diameter can be set from 30mm to 120mm, the effective light transmission diameter of a single filter can be set from 10mm to 50mm, and the wheel thickness can be set from 5mm to 25mm. The transmittance levels between the filters can be increased logarithmically or proportionally to cover different light emission states from weak to strong light.

[0086] When the system starts, the synchronous control module, after completing the preparation of the detection station, selects an appropriate filter level based on the brightness level of the MicroLED array under test and the testing task, driving the filter wheel to rotate to the corresponding position. This ensures that the light output from the conical lens undergoes predetermined attenuation or band filtering before entering the camera. As the MicroLED array on the sample stage emits light synchronously under the action of the electrical drive module, the conical lens maps the emitted light from different angles onto different areas of the image plane. After passing through the filter wheel, the light intensity entering the camera is adjusted to a suitable range, allowing the camera to acquire a light-emitting image that is neither too overexposed nor too dark. When the system detects a high-brightness MicroLED array, the filter wheel switches to a lower transmittance level (e.g., ND3.0) to reduce the light intensity entering the camera to within the camera's linear response range. When the system detects a low-brightness sample, the wheel switches to a higher transmittance level (e.g., ND0.3) to retain more effective signals and improve image resolution. The image correction module then processes the image using preset vignetting compensation and geometric distortion correction parameters. In this way, the filter wheel, in conjunction with the conical lens and camera, can perform matched exposure on samples of different brightness while keeping the conical imaging angle mapping relationship unchanged. This expands the measurement dynamic range, reduces the impact of saturation distortion or insufficient signal-to-noise ratio on the detection results, improves the adaptability of the same detection device to different MicroLED samples, different luminescence states and different process batches, and enhances the repeatability and versatility of the full field of view measurement results.

[0087] In one possible embodiment, the image correction module is deployed in the main control computer as a core component of the measurement algorithm software. It performs vignetting compensation and geometric distortion correction on the luminous images acquired by the camera. The main control computer is equipped with a high-performance GPU (such as an NVIDIA RTX 4090) for parallel image processing and connects to an image acquisition card via a high-speed Peripheral Component Interconnect Express (PCIe) interface. The image acquisition card supports Camera Link HS or CoaXPress interfaces, with a data transfer rate ≥25Gbps, ensuring real-time transmission of 14-bit image data to the main control computer. The measurement algorithm software runs on the main control computer and includes functional modules such as system control, image acquisition, correction algorithm, data analysis, and report generation. The correction algorithm is the specific implementation of the image correction module.

[0088] The system employs strict timing control to ensure measurement accuracy. For example, the main control computer initiates the measurement via software at time T0. The synchronization control module (e.g., FPGA) receives the start signal at time T0+100μs and begins initialization. Subsequently, at time T0+200μs, the synchronization control module sends an enable signal to the multi-channel programmable power supply. At time T0+300μs, the multi-channel programmable power supply is established, and the MicroLED array begins to emit light. At time T0+310μs, the synchronization control module sends a trigger signal to the camera (exposure begins). At time T0+310μs + exposure time (typically 5ms), the camera exposure ends, and image data transmission begins. At time T0+10ms, the power supply is turned off to reduce LED self-heating. The jitter of the entire synchronization timing is less than 1μs, ensuring that all 1500+ MicroLEDs are measured under the same electrical conditions.

[0089] The synchronization control module communicates with the host computer via USB 3.0 or Ethernet to receive measurement parameter configurations. Its output signals include the camera trigger signal (TTL level, rising edge trigger, jitter <1μs) and the power synchronization signal (controlling the output timing of the multi-channel programmable power supply). The multi-channel programmable power supply connects to the host computer via GPIB or LAN interface to receive current setpoints and switching commands. This power supply has 32 independent output channels, each with an output current range of 0.1mA-100mA, a resolution of 16 bits, a settling time of <10μs, and connects to the MicroLED array via a B2B electrical connection interface. The gating control circuit connects to the host computer via a GPIO interface and to the B2B electrical connection interface via a row and column gating circuit to achieve row and column gating control.

[0090] Accordingly, the image correction module adopts a "calibrate first, then correct" strategy. Specifically, the vignetting coefficient matrix and geometric distortion model are calibrated in advance. Then, following the order of "vignetting compensation first, then geometric distortion correction," vignetting compensation is performed on the luminous image based on the pre-calibrated vignetting coefficient matrix. First, the brightness attenuation at the edge of the field of view is restored, making the pixel grayscale values ​​consistent with the actual light intensity distribution, resulting in a vignetted-compensated image. Then, based on the pre-calibrated geometric distortion model, geometric distortion correction is performed on the vignetted-compensated image to eliminate the nonlinear mapping deviation of the conical lens, making the pixel positions correspond to the actual angles, resulting in a corrected image. Further, the emission angle distribution and optical parameters of each MicroLED in the MicroLED array are extracted from the corrected image. If the order is reversed, i.e., "geometric distortion correction first, then vignetting compensation," geometric distortion correction may lead to distortion of the brightness distribution after vignetting compensation, thereby reducing the accuracy of angle distribution extraction.

[0091] The above embodiments illustrate the implementation of the micro LED luminescence characteristic detection device. Next, specific embodiments will be used to describe the application of this device.

[0092] Figure 4 This is a schematic flowchart of an image correction method provided for an exemplary embodiment of this application. The image correction method provided in this application embodiment is applied to the image correction module in a micro LED luminescence characteristic detection device as described in any of the above embodiments. Figure 4 As shown, the image correction method includes:

[0093] S401. Obtain the light emission image of the MicroLED array.

[0094] For example, the main control computer sends a trigger signal to the camera through the synchronization control module. Under the control of the trigger signal, the camera exposes and acquires the image formed by the conical lens. The conical lens is set between the camera and the sample stage to map light emitted from the MicroLED array at different angles to different positions on the image plane. The camera receives the light during exposure and converts it into digital image data. The image acquisition card is connected to the camera through a high-speed data interface and transmits the acquired luminescent image to the main control computer in real time as the raw image data for subsequent correction processing.

[0095] S402. Based on the pre-calibrated vignetting coefficient matrix, perform vignetting compensation on the luminescent image to obtain the vignetting-compensated image.

[0096] In this embodiment, vignetting compensation refers to the position-dependent intensity correction of pixel grayscale values ​​in the original luminous image to address the brightness attenuation phenomenon at the edge of the field of view introduced by the conical lens, camera sensor, and system optical path, in order to restore the true light intensity distribution that should correspond to different image plane positions. The vignetting coefficient matrix is ​​a two-dimensional set of compensation parameters corresponding to the image pixel array, where each matrix element corresponds to a compensation factor for a pixel position or pixel region in the image, used to characterize the degree of transmittance attenuation at that position relative to a reference position. Pre-calibration refers to acquiring a calibration image before actual detection using a standard light source, standard uniform luminous surface, standard luminous sample, or system-level reference sample, under predetermined lens, camera, working distance, aperture parameters, and mounting posture, and generating a vignetting coefficient matrix based on the relationship between the theoretical brightness and measured brightness at each position in the calibration image. After receiving the luminous image obtained in S401, the image correction module first reads the vignetting coefficient matrix matching the current detection mode, and then performs compensation calculations point-by-point according to pixel coordinates to correct the distortion in the system transmittance difference between the image center region and the edge region.

[0097] S403. Based on the pre-calibrated geometric distortion model, perform geometric distortion correction on the image after vignetting compensation to obtain the corrected image.

[0098] In this embodiment, geometric distortion correction refers to the inversion and correction of the spatial nonlinear mapping relationship caused by the conical lens and imaging system, so as to establish an accurate correspondence between the pixel position in the image and the actual ideal angular position or ideal geometric position. The geometric distortion model is a mathematical model or mapping model used to describe the functional relationship between the actual imaging coordinates and the ideal coordinates. This model can be a polynomial model, a radial-tangential distortion model, a piecewise local model, or a lookup table model. For the detection of MicroLED luminous characteristics, the conical lens does not simply pursue conventional planar imaging fidelity, but needs to map the emitted light over a large angular range onto a finite image plane. Therefore, there is a nonlinear relationship between the pixel coordinates and the emission angle. Without geometric distortion correction, it is difficult to accurately recover the true luminous distribution of each MicroLED at different angles.

[0099] Specifically, the image correction module determines the offset relationship between the position of each pixel in the vignetting-compensated image and its ideal position based on the geometric distortion model. Through coordinate remapping, the distorted image is restored to a distortion-free image, thus obtaining the corrected image. This process effectively eliminates edge distortion introduced by large-angle conical lenses due to their ultra-wide viewing angle, restoring the linear mapping relationship between angle and pixel position.

[0100] S404. Extract the emission angle distribution and optical parameters of each MicroLED in the MicroLED array from the corrected image.

[0101] In this application, the emission angle distribution refers to the light intensity variation of a single MicroLED in different emission angle directions. It can be represented as a one-dimensional curve between angle and brightness, a polar coordinate distribution, a two-dimensional angle energy diagram, or other forms suitable for evaluating light emission characteristics. The optical parameters are quantitative indicators calculated based on this angle distribution and the corresponding pixel values ​​in the corrected image. These parameters may include peak brightness, center brightness, integrated light intensity, half-power angle, full width at half height angle, main emission direction, angle range, emission symmetry, uniformity, chromaticity-related parameters, and array dispersion, etc.

[0102] Specifically, the image correction module can first perform background residual removal and smoothing filtering on each MicroLED sub-region, then search for the peak light intensity position to determine the coordinates corresponding to the main emission direction and the central optical axis. If the correction coordinate system has been directly established on the scale of angles, the pixel energy in the sub-region can be accumulated according to the angle radius or angle grid to obtain the angle distribution curve; if the correction coordinate system is still a geometric coordinate, the pixel coordinates are converted into corresponding angle values ​​according to the preset angle mapping formula or mapping table, and then the intensity is statistically analyzed according to the angle interval. The half-power angle can be obtained by finding the left and right angle positions corresponding to the brightness dropping to half of the peak and calculating the included angle; the peak brightness can be obtained from the maximum pixel value or the locally fitted peak; the integrated light intensity can be obtained by summing or integrating all pixels within the effective angle range; the light emission uniformity can be calculated by comparing the brightness standard deviation, range, or normalized deviation of different angle regions. For array-level quality evaluation, after obtaining the parameters of each MicroLED, the image correction module can further generate parameter distribution maps, statistical histograms, defect lists, and process trend analysis results for the entire wafer or array, which can be used for production line screening and process feedback. If the detection needs to take into account the chromaticity parameters, the same partitioning and angle extraction process can be performed on different color channels under multi-band imaging conditions, and the color coordinates, dominant wavelength or related chromaticity index can be obtained by combining the color matching function or system calibration relationship.

[0103] In this embodiment, by organically linking image acquisition, light intensity compensation, spatial mapping correction, and parameter extraction in a predetermined sequence, rapid and accurate processing of large-angle emission information from a high-density MicroLED array is achieved. First, vignetting compensation restores the true brightness at the edge of the field of view; then, geometric distortion correction restores the correspondence between pixel positions and ideal angular positions. Finally, based on unified correction, the angular distribution and optical parameters of a single MicroLED are extracted, effectively improving the technical problems of slow batch detection, insufficient acquisition of full-angle information, and measurement distortion in edge regions in existing technologies. Based on the above processing chain, the system can improve the parallelism, accuracy, and consistency of high-density MicroLED array emission characteristic evaluation while maintaining synchronous emission detection of the array.

[0104] In some embodiments, the vignetting coefficient matrix is ​​determined as follows: acquiring a vignetting calibration image under uniform light illumination; extracting the average brightness of the central region of the vignetting calibration image as a reference brightness value; determining the ratio of the brightness value of each pixel in the vignetting calibration image to the reference brightness value to generate an initial vignetting coefficient matrix; and performing smoothing filtering on the initial vignetting coefficient matrix to obtain the vignetting coefficient matrix.

[0105] The vignetting calibration image under uniform light source illumination is used to characterize the brightness attenuation characteristics of the imaging system at different locations within the field of view. For example, an integrating sphere uniform light source illuminates the entire field of view, with a brightness uniformity greater than 99%, to ensure that the light intensity received at each location is as consistent as possible during calibration. After the camera captures the uniform light source image, the average brightness of the central region of the image (e.g., ±10% of the field of view) is extracted as the reference brightness value. This central region serves as a reference brightness benchmark to reduce the impact of edge attenuation on the benchmark value. Then, the ratio of the brightness value of each pixel in the vignetting calibration image to the benchmark brightness value is determined to generate an initial vignetting coefficient matrix. When the brightness of a pixel is lower than the benchmark brightness, this ratio is used to characterize the degree of vignetting attenuation at that location, thus forming a coefficient matrix consistent with the image size. Further, the initial vignetting coefficient matrix is ​​smoothed using a smoothing filter. This smoothing filter can employ Gaussian filtering (e.g., σ=5 pixels) or mean filtering to suppress local noise, isolated outliers, and discrete fluctuations caused by acquisition jitter, ensuring a continuous and stable coefficient distribution, facilitating subsequent pixel-level brightness compensation of the MicroLED luminescent image. Accordingly, the vignetting coefficient matrix can be stored as a vignetting look-up table (Vignetting LUT).

[0106] For example, the vignetting coefficient satisfies the following formula:

[0107]

[0108] in, The vignetting coefficient represents the position in the image. The required brightness compensation factor; The average brightness of the central region is usually extracted as the baseline value, which is the average brightness of the area ±10% of the field of view at the center of the image. The brightness of the image is determined by a uniform light source, specifically the position of the vignetting calibration image captured under uniform illumination from an integrating sphere. The pixel brightness value at that location.

[0109] In practical applications, after acquisition, the image correction module calls the pre-stored vignetting coefficient matrix to compensate the corresponding positions of the luminous image, restoring the brightness response of the edge region to a reference level consistent with the central region. Because the coefficient matrix is ​​smoothed, the compensation result reduces brightness steps and artifacts caused by abrupt changes in local coefficients, thereby improving the accuracy of optical parameter extraction from the array edge chips. This method can explicitly quantify the field-of-view non-uniformity of the lens and imaging channel and perform unified correction during the calibration phase, thus improving the consistency and repeatability of large-field-of-view MicroLED array detection.

[0110] Based on the above embodiments, in some embodiments, vignetting compensation is performed on the luminous image according to a pre-calibrated vignetting coefficient matrix, including: for each pixel in the luminous image, obtaining the vignetting coefficient corresponding to the pixel's position in the vignetting coefficient matrix, and using the product of the pixel's gray value and the vignetting coefficient as the compensated gray value corresponding to the pixel to obtain the vignetting-compensated image.

[0111] In one possible implementation, the image correction module performs vignetting compensation on the luminous image based on a pre-calibrated vignetting coefficient matrix. Specifically, for each frame of the acquired measured MicroLED luminous image, the image correction module retrieves a pre-stored vignetting coefficient lookup table, which stores the vignetting coefficient corresponding to each pixel position. For each pixel in the luminous image, the image correction module retrieves the vignetting coefficient corresponding to that pixel position from the vignetting coefficient lookup table, multiplies the pixel's grayscale value by the vignetting coefficient, and obtains the compensated grayscale value for that pixel. This restores the brightness attenuated in the edge area due to lens vignetting to the true level, resulting in a vignetting-compensated image.

[0112] For example, the compensated grayscale value satisfies the following formula:

[0113]

[0114] in, Location in the image The corresponding compensated grayscale value; In the luminescent image The corresponding grayscale value.

[0115] Furthermore, after vignetting compensation, the image correction module can also perform a bad pixel correction step. Specifically, it detects hot pixels and dead pixels in the camera sensor. Hot pixels are abnormal pixels that consistently exhibit high brightness values, while dead pixels are abnormal pixels that consistently exhibit low brightness values ​​or zero brightness values. For detected bad pixels, the image correction module uses interpolation methods (such as neighborhood pixel mean interpolation or median interpolation) to repair them, in order to eliminate the interference of bad pixels on subsequent geometric distortion correction and parameter extraction.

[0116] Through the above-mentioned vignetting compensation and bad pixel correction processing, the inherent edge brightness attenuation of large-angle conical lenses and the pixel defects of the camera sensor itself can be effectively eliminated, ensuring the brightness consistency and data integrity of the luminescent image across the entire field of view, thereby providing high-quality image data for subsequent geometric distortion correction and luminescence characteristic analysis.

[0117] In some embodiments, the geometric distortion model is determined as follows: the imaging position of a single MicroLED at different angles is obtained, and a mapping relationship between the actual angle and the radial distance of the image plane is established; the theoretical radial distance corresponding to each angle is determined according to the theoretical mapping relationship when there is no distortion; the deviation between the actual radial distance and the theoretical radial distance is determined to obtain the distortion deviation; and the relationship between the distortion deviation and the theoretical radial distance is fitted using a polynomial to obtain the geometric distortion model.

[0118] In practical implementation, a precision rotary stage (e.g., with an angular resolution of 0.01°) is used in conjunction with a single standard MicroLED. The standard MicroLED is placed at the center of the field of view, and its imaging position at different angles is recorded, for example, in 1° increments within the range of 0° to 70°. The radial distance of the image plane refers to the radial offset of the imaging position relative to the image center. Through the above acquisition, the actual angle can be established. The mapping relationship with the radial distance of the image plane, i.e. Due to distortion, the actual measured radial distance is... There is a discrepancy between the calculated radial distance and the theoretical radial distance based on an ideal, distortion-free optical path (such as a pinhole imaging model), i.e. The distortion deviation can be obtained by determining the difference between the actual radial distance and the theoretical radial distance. ,Right now .

[0119] Accordingly, to reduce the impact of sampling noise on model stability, a 9th-order polynomial is used to fit the relationship between distortion bias and theoretical radial distance. The resulting distortion curve satisfies the following formula:

[0120]

[0121] in, These are the distortion coefficients. This polynomial is the geometric distortion model, used to describe the nonlinear offset of the actual imaging position relative to the theoretical imaging position. The odd-order design of the polynomial ensures the symmetry of the geometric distortion model about the optical axis. After fitting, the distortion coefficients and the inverse mapping lookup table (i.e., the distortion lookup table) are stored for use in the subsequent online correction stage.

[0122] In addition, brightness linearity calibration was performed. Specifically, a standard luminance meter (such as Konica Minolta CS-2000) was used to calibrate different brightness levels, establish the mapping relationship between camera grayscale values ​​and actual brightness, ensure the absolute accuracy of 14-bit grayscale values, and thus provide an accurate brightness benchmark for subsequent quantitative analysis of luminous intensity parameters.

[0123] By adopting the above method, the geometric distortion model can simultaneously take into account theoretical geometric relationships and actual optical system errors, accurately characterizing imaging shifts at different angles, thereby improving the spatial consistency and angular distribution restoration accuracy of large-field-of-view MicroLED luminescent images. Due to the use of polynomial fitting, the model has good continuity and computability, facilitating rapid invocation in the image correction module, thus improving correction efficiency and comparability of detection results in batch detection scenarios.

[0124] Based on the above embodiments, in some embodiments, geometric distortion correction is performed on the vignetting-compensated image according to a pre-calibrated geometric distortion model, including: for each pixel coordinate in the corrected image, determining the normalized coordinates of the pixel coordinates relative to the center of the corrected image, and determining the theoretical radial distance based on the normalized coordinates; obtaining the distortion deviation corresponding to the theoretical radial distance according to the geometric distortion model, and determining the actual radial distance based on the distortion deviation; determining the corresponding source coordinates of the pixel coordinates in the vignetting-compensated image based on the ratio of the actual radial distance to the theoretical radial distance; and sampling the pixel value at the corresponding source coordinates from the vignetting-compensated image as the grayscale value of the pixel coordinates in the corrected image.

[0125] In its implementation, the image correction module uses a coordinate remapping method based on a distortion lookup table. For each pixel coordinate in the corrected image... First, calculate the normalized coordinates of the pixel relative to the image center. The normalized coordinates satisfy the following formula:

[0126]

[0127] In the formula, The coordinates of the principal point (i.e., the image center) are given. and This represents the equivalent focal length of the conical lens. Normalized coordinates are used to eliminate the influence of image resolution differences on coordinate mapping, thus obtaining a unified radial analysis benchmark. Subsequently, the theoretical radial distance is determined based on the normalized coordinates. and ideal angle :

[0128]

[0129] The theoretical radial distance characterizes the geometric distance of a target pixel from the image center under ideal distortion-free imaging conditions. Subsequently, the actual radial distance is determined based on a pre-calibrated geometric distortion model. Specifically, it satisfies the following formula:

[0130]

[0131] for The corresponding distortion deviation, calculated using a 9th-order polynomial, describes the offset of the actual imaging position relative to the theoretical position. Next, the scaling factor is determined, and based on this, the corresponding source coordinates of the pixel in the vignetting-compensated image are determined. :

[0132]

[0133]

[0134] in, This is the scaling factor.

[0135] The above correction method employs an inverse mapping strategy, which calculates the corresponding position in the original image from the corrected image coordinates, thereby reducing the holes and redundant assignments caused by forward resampling. Subsequently, the pixel value at the corresponding source coordinate is sampled from the vignetting-compensated image. Bicubic interpolation is used for sampling to ensure the continuity and edge smoothness of the reconstructed image, and the sampled pixel value is used as the grayscale value of that pixel coordinate in the corrected image.

[0136] Because the correction process uses the image center as a reference and corrects according to radial characteristics, it can effectively eliminate the nonlinear spatial offset introduced by the conical lens, ensuring that the MicroLED emission angle distribution maintains a true correspondence in the corrected image. This implementation further eliminates the influence of geometric distortion on pixel positions in edge regions based on vignetting compensation, significantly reducing angle measurement errors and thus significantly improving the spatial consistency and angle mapping accuracy of large field-of-view imaging results. This enhances the reliability of MicroLED array emission characteristic extraction, providing a stable and accurate image basis for subsequent optical parameter extraction.

[0137] Further, after completing geometric distortion correction, the image correction module extracts angle-brightness information. Specifically, for the imaging area of ​​each MicroLED, the Region of Interest (ROI) is determined based on pre-calibrated position information, and the corresponding conical image sub-region of the chip is extracted from the corrected image, typically 50×50 pixels in size. Ambient light background is subtracted; this background can be obtained by pre-capturing a dark-field image without LED emission to eliminate interference from ambient light on the measurement results. Next, angle distribution calculation is performed: for each pixel in the ROI, based on its position relative to the chip center... The corresponding emission angle is calculated, specifically satisfying the following formula:

[0138]

[0139]

[0140] in, The angle between the emitted ray and the vertical direction (optical axis) indicates the degree of tilt of the ray; The angle between the projection of the emitted ray onto the horizontal plane and a reference direction (such as the X-axis) indicates the direction of the ray. This represents the offset distance of a pixel relative to the chip center along the X-axis. This represents the offset distance of a pixel relative to the chip center along the Y-axis. Let be the equivalent focal length of the conical lens. Accordingly, construct the angle-luminance distribution function. Furthermore, optical parameters are calculated, specifically including peak brightness, half-power angle, luminous uniformity, and total luminous flux, which satisfy the following formulas:

[0141]

[0142]

[0143]

[0144]

[0145] in, Peak brightness, which is the maximum value of the angle-brightness distribution; It is the half-power angle, which satisfies Angular range; luminous uniformity , which is the ratio of minimum brightness to maximum brightness; total luminous flux. This refers to the solid angle integral of the angle-brightness distribution. Through the above processing, the image correction module can output the complete emission angle distribution and key optical parameters of each MicroLED, thereby providing a quantitative basis for the optical performance evaluation and defect determination of the MicroLED array.

[0146] To make the technical solution of this application clearer, the following describes this application in further detail with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only for illustrating this application and do not constitute any limitation on this application.

[0147] Example 1: 4-inch wafer-level MicroLED detection

[0148] This embodiment uses Figure 3 Taking the schematic diagram as an example, the micro-LED luminescence characteristic detection device provided in the above embodiment is used to detect a 4-inch GaN-based blue MicroLED wafer. Approximately 2000 MicroLEDs are arrayed on the wafer, with a chip size of 30μm × 30μm and a chip spacing of 50μm.

[0149] The hardware configuration in this embodiment is as follows: the conical lens has a viewing angle of ±75°, a working distance of 10mm, and an equivalent focal length of f=8mm; the camera is a 14-bit scientific-grade complementary metal-oxide-semiconductor camera with a resolution of 2048×2048 pixels and a pixel size of 6.5μm; the XYZ three-axis translational travel of the five-axis carrier platform is 100mm×100mm×20mm, and the positioning accuracy is 0.5μm; the multi-channel programmable power supply uses a 32-channel precision constant current source with a resolution of 16bit and a maximum output current of 50mA per channel; the synchronization control module uses a field-programmable gate array (FPGA) controller with a trigger jitter of less than 500ns.

[0150] The specific measurement process is as follows: Step 1: Load the MicroLED wafer onto the sample stage and fix it on the stage using a vacuum adsorption mechanism. Step 2: Perform coarse alignment by moving the wafer to the center of the field of view using a low-magnification camera. Step 3: Perform fine alignment by imaging with a conical lens and adjusting the X and Y axes to center the target chip array. Step 4: Perform leveling by adjusting the θx and θy axes to make the wafer surface perpendicular to the optical axis of the conical lens. The leveling criterion is to monitor the sharpness of the edge chips. Step 5: Perform B2B electrical connection interface docking by lowering the sample stage to dock the flexible printed circuit board with the B2B electrical connection interface. Step 6: Call the vignetting coefficient lookup table and distortion lookup table pre-stored during the system calibration stage. Step 7: Perform synchronous measurement by setting the exposure time to 5ms and the drive current to 10mA, with synchronous triggering executed by the synchronous control module. Step 8: Perform image processing by using a graphics processor for parallel image correction and extracting the optical parameters of 1500 chips. Step 9: Output the detection results, generating a brightness distribution map, an angle distribution statistical map, and a defect marking map.

[0151] The measurement results of this embodiment are as follows: the number of chips covered in a single measurement is 1500 (effective area); the measurement time is 5ms for image acquisition, 50ms for data transmission, 100ms for image correction, and 500ms for data analysis, totaling 655ms; the brightness repeatability is ±1.5% (3σ); the angle measurement accuracy is ±0.5° (within the range of 0° to 60°); the defect detection rate is greater than 99%, and the detected objects include dead spots, bright spots, and uneven brightness defects.

[0152] The calibration effect was verified as follows: Without calibration, the brightness measurement value of the edge chip (angle greater than 60°) was 35% lower than that of the center area; after vignetting compensation, the edge brightness recovered to within ±2% of the true value; without calibration, the edge angle measurement error reached 2.5°; after geometric distortion correction, the full field of view angle error was less than 0.5°.

[0153] Example 2: Simultaneous detection of red, green, and blue (RGB) microLEDs

[0154] This embodiment uses an RGB three-color MicroLED array as the detection object and employs the same micro-LED luminescence characteristic detection device as in Embodiment 1. The luminescence characteristics of different color chips are measured through three exposures.

[0155] The specific measurement procedure is as follows: First exposure: Only the red MicroLED is lit, while the green and blue MicroLEDs remain off. The angle-brightness distribution of the red light is measured. Second exposure: Only the green MicroLED is lit, while the red and blue MicroLEDs remain off. The angle-brightness distribution of the green light is measured. At this time, a neutral density filter is installed between the conical lens and the camera to prevent high-brightness green light from causing camera pixel saturation. Third exposure: Only the blue MicroLED is lit, while the red and green MicroLEDs remain off. The angle-brightness distribution of the blue light is measured. The above three measurements use the same pre-calibrated vignetting coefficient lookup table and distortion lookup table, based on the assumption that the color difference of the conical lens is negligible. After the measurements are completed, optical performance distribution maps for red, green, and blue are generated, including parameters such as the brightness distribution, angle distribution, and uniformity of each color chip.

[0156] In this embodiment, since the camera is a 14-bit high dynamic range camera with a dynamic range of over 70dB, it can simultaneously and accurately measure MicroLEDs of different brightness levels, from low-brightness red light (less than 100cd / m²) to high-brightness green light (greater than 100,000cd / m²). This effectively reduces the time overhead and synchronization problems caused by multiple exposures or gain switching, thereby significantly improving the detection efficiency and measurement consistency of the RGB three-color MicroLED array.

[0157] In summary, this application has at least the following advantages:

[0158] The traditional method of rotating and measuring one microLED at a time to inspect 1500 microLEDs takes about 3 hours. However, this application uses a synchronous control module to coordinate the electrical drive module and the optical acquisition module in a timely manner, and uses a parallel electrical connection interface and a multi-channel programmable power supply to drive the microLED array to emit light simultaneously. Combined with a conical lens for single-exposure imaging, the single measurement cycle is less than 1 second, which can simultaneously complete the inspection of multiple microLEDs. The time to inspect more than 1500 microLEDs is reduced from hours to seconds, and the inspection efficiency is improved by more than 10,000 times. This can better meet the cycle time requirements of mass inspection at the production line level, so that inspection is no longer a bottleneck for production capacity.

[0159] Second, by employing a large-angle conical lens (viewing angle ≥ ±70°), compared to traditional microscope lenses (viewing angle only ±30°), the angle collection range is expanded by more than 2.5 times, enabling complete capture of the large-angle emission signals of MicroLEDs. Combined with vignetting compensation and geometric distortion correction algorithms, the complete emission angle distribution of each MicroLED can be extracted from a single frame image. This provides key angular characteristic data that traditional methods cannot obtain for near-eye display applications such as augmented reality / virtual reality (AR / VR), significantly improving the accuracy of display uniformity and viewing experience evaluation. By using a 14-bit high dynamic range camera with a dynamic range of over 70dB, coupled with precise exposure control and synchronous triggering timing, MicroLEDs at different brightness levels can be accurately measured simultaneously in a single exposure, from low-brightness red light (less than 100 cd / m²) to high-brightness green light (greater than 100,000 cd / m²). This effectively reduces the time overhead and synchronization problems caused by multiple exposures or gain switching, thereby effectively improving the detection efficiency and measurement consistency of RGB three-color MicroLED arrays.

[0160] Third, through vignetting compensation and geometric distortion correction algorithms, the inherent imaging defects of large-angle conic lenses are systematically corrected. Experimental verification shows that, without correction, the brightness measurement error in the edge region (angle greater than 60°) can reach more than 35%, and the angle measurement error can reach more than 2.5°. After vignetting compensation, the edge brightness is restored to within ±2% of the true value; after distortion correction, the angle error across the entire field of view is less than 0.5°. Compared with the uncorrected conic imaging system, the measurement accuracy is improved by an order of magnitude, effectively ensuring the accuracy of photometric measurements across the entire field of view.

[0161] Fourth, by designing a five-axis carrier platform compatible with standard wafer sizes (4-inch, 6-inch, and 8-inch), precise positioning and leveling of MicroLED wafers can be achieved. By using a B2B electrical connection interface instead of traditional probe cards, the risk of damage from probe-contact measurements is reduced, significantly improving connection repeatability and test stability. Simultaneously, the device described in this application can seamlessly interface with upstream and downstream process equipment (such as mass transfer equipment and sorting equipment), facilitating a fully automated inspection process and effectively reducing manual intervention and operational errors. Furthermore, it significantly reduces equipment costs and maintenance costs. In terms of production capacity, inspection is no longer a bottleneck, supporting the inspection needs of large-scale panels. Regarding yield, rapid full inspection can promptly identify process problems, thereby effectively improving the production yield of MicroLEDs.

[0162] Based on the above-mentioned technical advantages, this application effectively improves the technical problem that existing detection schemes are unable to achieve rapid, synchronous, and high-precision optical performance detection of multiple MicroLEDs, and realizes efficient, synchronous, and accurate detection of the light-emitting characteristics of MicroLED arrays, which has important industrial application value.

[0163] Figure 5 A schematic diagram of an image correction apparatus provided for an exemplary embodiment of this application. (See attached diagram.) Figure 5 As shown, the image correction device 50 includes an acquisition module 51, a vignetting compensation module 52, a geometric distortion correction module 53, and a parameter extraction module 54, wherein:

[0164] Acquisition module 51 is used to acquire the light emission image of the MicroLED array;

[0165] The vignetting compensation module 52 is used to perform vignetting compensation on the luminescent image according to the pre-calibrated vignetting coefficient matrix to obtain the vignetting compensated image;

[0166] The geometric distortion correction module 53 is used to perform geometric distortion correction on the image after vignetting compensation according to the pre-calibrated geometric distortion model, so as to obtain the corrected image.

[0167] The parameter extraction module 54 is used to extract the emission angle distribution and optical parameters of each MicroLED in the MicroLED array from the corrected image.

[0168] In one possible implementation, the vignetting coefficient matrix is ​​determined as follows: acquiring a vignetting calibration image under uniform light source illumination; extracting the average brightness of the central region of the vignetting calibration image as a reference brightness value; determining the ratio of the brightness value of each pixel in the vignetting calibration image to the reference brightness value to generate an initial vignetting coefficient matrix; and performing smoothing filtering on the initial vignetting coefficient matrix to obtain the vignetting coefficient matrix.

[0169] In one possible implementation, the vignetting compensation module 52 can be specifically used to: for each pixel in the luminous image, obtain the vignetting coefficient corresponding to the pixel's position in the vignetting coefficient matrix, and use the product of the pixel's gray value and the vignetting coefficient as the compensated gray value corresponding to the pixel to obtain the vignetting compensated image.

[0170] In one possible implementation, the geometric distortion model is determined as follows: the imaging position of a single MicroLED at different angles is obtained, and a mapping relationship between the actual angle and the radial distance of the image plane is established; the theoretical radial distance corresponding to each angle is determined based on the theoretical mapping relationship when there is no distortion; the deviation between the actual radial distance and the theoretical radial distance is determined to obtain the distortion deviation; and the relationship between the distortion deviation and the theoretical radial distance is fitted using a polynomial to obtain the geometric distortion model.

[0171] In one possible implementation, the geometric distortion correction module 53 can be specifically used to: for each pixel coordinate in the corrected image, determine the normalized coordinates of the pixel coordinates relative to the center of the corrected image, and determine the theoretical radial distance based on the normalized coordinates; obtain the distortion deviation corresponding to the theoretical radial distance based on the geometric distortion model, and determine the actual radial distance based on the distortion deviation; determine the corresponding source coordinates of the pixel coordinates in the vignetting-compensated image based on the ratio of the actual radial distance to the theoretical radial distance; and sample the pixel value at the corresponding source coordinates from the vignetting-compensated image as the grayscale value of the pixel coordinates in the corrected image.

[0172] The image correction device provided in this application embodiment can execute the technical solution shown in the above image correction method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0173] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented in software via processing element calls, while others are implemented in hardware. For example, the vignetting compensation module can be a separate processing element, or it can be integrated into a chip in the above device. Alternatively, it can be stored as program code in the memory of the above device, and its function can be called and executed by a processing element. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through the integrated logic circuits in the hardware of the processor element or through software instructions.

[0174] For example, these modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). As another example, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a System-On-a-Chip (SOC).

[0175] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Video Discs, DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0176] Figure 6 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application. For example... Figure 6 As shown, the electronic device 60 in this embodiment includes:

[0177] At least one processor 61; and a memory 62 communicatively connected to the at least one processor;

[0178] The memory 62 stores instructions that can be executed by at least one processor 61 to cause the electronic device to perform the method as described in any of the above embodiments.

[0179] Alternatively, the memory 62 can be either standalone or integrated with the processor 61.

[0180] The memory 62 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0181] The processor 61 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. Specifically, when implementing the image correction method described in the foregoing method embodiments, the electronic device may be, for example, an electronic device with processing capabilities such as a server.

[0182] Optionally, the electronic device may also include a communication interface 63. In specific implementations, if the communication interface 63, memory 62, and processor 61 are implemented independently, they can be interconnected via a bus to complete communication. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc., but this does not imply that there is only one bus or one type of bus.

[0183] Optionally, in a specific implementation, if the communication interface 63, memory 62, and processor 61 are integrated on a single chip, then the communication interface 63, memory 62, and processor 61 can communicate through an internal interface.

[0184] The implementation principle and technical effects of the electronic device provided in this embodiment can be found in the foregoing embodiments, and will not be repeated here.

[0185] This application also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are executed, they are used to implement the method steps as described in the above method embodiments. The specific implementation methods and technical effects are similar and will not be repeated here.

[0186] The aforementioned computer-readable storage media can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0187] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in an image correction device.

[0188] This application also provides a computer program product, including a computer program, which, when executed, implements the method steps as described in the above method embodiments. The specific implementation and technical effects are similar and will not be repeated here.

[0189] This application also provides a micro-LED production line-level optical performance testing system, including the micro-LED luminescence characteristic testing device of any of the above embodiments, wherein the single measurement cycle is less than 1 second, and more than 1,500 MicroLEDs can be tested simultaneously in one measurement.

[0190] In practical applications, this testing system is integrated into the MicroLED mass production line, seamlessly connecting with upstream and downstream process equipment (such as mass transfer equipment and sorting equipment) to achieve a fully automated testing process. During operation, the MicroLED wafers to be tested are automatically loaded onto the sample stage by an automated loading and unloading mechanism. The five-axis carrier platform completes precise positioning and leveling, the B2B electrical connection interface automatically connects, and the synchronous control module executes synchronous triggering. The single measurement cycle is less than 1 second, and more than 1,500 MicroLEDs can be tested simultaneously at one time.

[0191] This inspection system is compatible with standard 4-inch, 6-inch, and 8-inch wafer sizes and is suitable for mass optical inspection of MicroLED wafers, mass transfer arrays, and packaged devices. After inspection, the system automatically generates brightness distribution maps, angle distribution statistics maps, and defect marking maps, and uploads the inspection data to the production line management system, providing data support for process optimization and yield improvement.

[0192] This inspection system, employing the aforementioned micro-LED luminescence characteristic detection device, achieves efficient, synchronous, and precise inspection of MicroLED arrays. A single measurement cycle is less than 1 second, and over 1500 MicroLEDs can be inspected at once. The inspection efficiency is more than 10,000 times higher than traditional serial measurement methods, fully meeting the cycle time requirements of mass production line inspections. Furthermore, replacing traditional probe cards with a B2B electrical connection interface significantly reduces the risk of damage from contact measurements, improving connection repeatability and test stability. In addition, the synergistic effect of the vacuum adsorption mechanism and the five-axis carrier platform ensures the positional stability and repeatability of the wafers during the inspection process. This system effectively supports the production line-level inspection needs of large-scale MicroLED panels, significantly improving inspection efficiency and yield, and possesses significant industrial application value.

[0193] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0194] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0195] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A device for detecting the luminous characteristics of a micro-light-emitting diode, characterized in that, include: The synchronization control module is used to generate enable and trigger signals; The electrical driving module includes a sample stage for carrying a MicroLED wafer, a parallel electrical connection interface integrated on the sample stage, and a multi-channel programmable power supply electrically connected to the parallel electrical connection interface. The electrical driving module is electrically connected to the synchronization control module and is used to drive the MicroLED array to emit light simultaneously in response to the enable signal. An optical acquisition module includes a conical lens and a camera optically coupled to the conical lens. The optical acquisition module is electrically connected to the synchronization control module and is used to acquire the emitted image of the MicroLED array in response to a trigger signal. The conical lens is used to map light rays at different angles to different positions of the emitted image; wherein the viewing angle of the conical lens is ≥ ; and / or, the camera has a bit depth ≥ 14 bits and a dynamic range ≥ 70 dB; An image correction module, electrically connected to the camera, is used to perform vignetting compensation and geometric distortion correction on the luminous image to restore the correspondence between pixel brightness and actual angle. This includes: performing vignetting compensation on the luminous image according to a pre-calibrated vignetting coefficient matrix to obtain a vignetted-compensated image; and then performing geometric distortion correction on the vignetted-compensated image according to a pre-calibrated geometric distortion model to obtain a corrected image. The synchronization control module includes a field-programmable gate array (FPGA), which is used for: The enable signal is output to the multi-channel programmable power supply to control the multi-channel programmable power supply to provide drive current during the illumination of the MicroLED array; The trigger signal is output to the camera to control the camera's exposure during the illumination of the MicroLED array; the time interval between the enable signal and the trigger signal is 100μs to 500μs, and the output jitter of the trigger signal is less than 1μs; The electrical drive module further includes a five-axis carrier platform, on which the sample stage is mounted. The five-axis carrier platform includes: an XYZ three-axis translation module for adjusting the position of the MicroLED array in the horizontal and vertical directions; and an θx-θy two-axis tilt adjustment module for adjusting the tilt angle of the MicroLED array relative to the optical axis of the conical lens. The vignetting coefficient matrix is ​​determined as follows: a vignetting calibration image under uniform light source illumination is acquired; the average brightness of the central region of the vignetting calibration image is extracted as a reference brightness value; the ratio of the brightness value of each pixel in the vignetting calibration image to the reference brightness value is determined to generate an initial vignetting coefficient matrix; and the initial vignetting coefficient matrix is ​​subjected to smoothing filtering to obtain the vignetting coefficient matrix. The geometric distortion model is determined as follows: the imaging position of a single MicroLED at different angles is obtained, and a mapping relationship between the actual angle and the radial distance of the image plane is established; based on the theoretical mapping relationship when there is no distortion, the theoretical radial distance corresponding to each angle is determined; the deviation between the actual radial distance and the theoretical radial distance is determined to obtain the distortion deviation; the relationship between the distortion deviation and the theoretical radial distance is fitted using a polynomial to obtain the geometric distortion model.

2. The micro-LED luminescence characteristic detection device according to claim 1, characterized in that, The parallel electrical connection interface is a board-to-board (B2B) electrical connection interface; And / or, the sample stage has a vacuum adsorption mechanism for fixing the MicroLED wafer.

3. The micro-LED luminescence characteristic detection device according to any one of claims 1 to 2, characterized in that, It also includes a filter wheel, which is disposed between the conical lens and the camera to extend the measurement dynamic range.

4. An image correction method, characterized in that, An image correction module applied in the micro-LED luminescence characteristic detection device as described in any one of claims 1 to 3, wherein the image correction method comprises: Acquire the emission image of the MicroLED array; According to the pre-calibrated vignetting coefficient matrix, the luminescent image is vignetted to obtain the vignetted compensated image; Based on a pre-calibrated geometric distortion model, the vignetting compensated image is subjected to geometric distortion correction to obtain the corrected image; The emission angle distribution and optical parameters of each MicroLED in the MicroLED array are extracted from the corrected image.

5. The image correction method according to claim 4, characterized in that, The step of performing vignetting compensation on the luminescent image according to a pre-calibrated vignetting coefficient matrix includes: For each pixel in the luminescent image, the vignetting coefficient corresponding to the position of the pixel in the vignetting coefficient matrix is ​​obtained, and the product of the gray value of the pixel in the luminescent image and the vignetting coefficient is used as the compensated gray value of the pixel to obtain the vignetting compensated image.

6. The image correction method according to claim 4, characterized in that, The step of performing geometric distortion correction on the vignetting-compensated image according to a pre-calibrated geometric distortion model includes: For each pixel coordinate in the corrected image, the normalized coordinates of the pixel coordinates relative to the center of the corrected image are determined, and the theoretical radial distance is determined based on the normalized coordinates; The distortion deviation corresponding to the theoretical radial distance is obtained based on the geometric distortion model, and the actual radial distance is determined based on the distortion deviation. Based on the ratio of the actual radial distance to the theoretical radial distance, the corresponding source coordinates of the pixel coordinates in the vignetting-compensated image are determined; The pixel value at the corresponding source coordinate is sampled from the image after vignetting compensation and used as the grayscale value of the pixel coordinate in the corrected image.

7. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory is used to store computer-executed instructions; The processor is configured to execute the computer execution instructions to implement the method as described in any one of claims 4-6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the method as described in any one of claims 4-6.

9. A micro-light-emitting diode production line-level optical performance testing system, characterized in that, The device includes the micro-light-emitting diode luminescence characteristic detection device as described in any one of claims 1 to 3, wherein the single measurement cycle is less than 1 second, and more than 1,500 micro-light-emitting diodes (MicroLEDs) can be detected simultaneously in a single measurement.

Images