Test method, system, electronic device, and storage medium for chip light emitting diode
By acquiring the RGB and light intensity images of the surface-mount LEDs under different power levels, aligning them, and performing image processing, the problem of ambient light affecting detection accuracy is solved, achieving highly accurate quality judgment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG YONGYU OPTOELECTRONICS CO LTD
- Filing Date
- 2023-11-09
- Publication Date
- 2026-07-21
AI Technical Summary
The quality inspection of existing surface-mount LEDs is easily affected by ambient light conditions, leading to inaccurate observation and misjudgment.
RGB images and light intensity images of the surface-mount LEDs (SLEDs) arrayed on the testing stage are acquired under different power levels. Image alignment and processing are performed, and the quality of the SLEDs is determined by combining the image processing results, thus avoiding manual inspection.
This improves the accuracy of surface-mount LED detection and reduces the impact of ambient light on the detection results.
Smart Images
Figure CN117572183B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of light-emitting diode testing, and more particularly to a testing method, system, electronic device, and storage medium for surface-mount light-emitting diodes. Background Technology
[0002] Surface mount LEDs (SMDs) are common optoelectronic components widely used in various electronic products. During the manufacturing process, the ability to quickly and accurately inspect the quality of SMDs is of paramount importance. Current methods primarily utilize ammeters and voltmeters for electrical parameter testing, and visual inspection to assess the light emission quality. This mainly involves visually observing the LED's emission pattern; uniform and bright light indicates good quality, while uneven or absent light suggests defects or damage. However, observing SMD LED emission is easily affected by ambient light conditions, leading to inaccurate observations, misjudgments, and ultimately, compromised quality upon delivery. Summary of the Invention
[0003] This invention provides a testing method for surface-mount LEDs (SMD LEDs), aiming to solve the problem that existing SMD LED quality inspection methods are easily affected by ambient light conditions when observing the light emission of SMD LEDs, leading to inaccurate observations, misjudgments, and affecting the factory quality of SMD LEDs. The method acquires the RGB and light intensity images of the SMD LEDs under test, arrayed on a testing platform, at different power levels. The RGB and light intensity images are aligned, and image processing is performed on the aligned RGB and light intensity images to obtain image processing results for different power levels. The method determines whether the SMD LED under test has passed the test based on these image processing results. This eliminates the need for manual observation of the light emission of the SMD LEDs under test. Furthermore, because the SMD LEDs under test are arrayed on the testing platform, the large light-emitting area minimizes the impact of environmental factors on the information acquired from the RGB and light intensity images, thus improving accuracy.
[0004] In a first aspect, embodiments of the present invention provide a testing method for surface-mount light-emitting diodes (SLEDs), the method comprising:
[0005] A forward-biased LED array is arranged on a testing platform, wherein the testing platform is provided with a testing station for the LED array and the testing station is provided with a forward power supply.
[0006] Under different power levels, the RGB image and light intensity image of the LED under test were obtained respectively.
[0007] Align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and light intensity image under the corresponding power level.
[0008] Image processing is performed on the image group to be processed under different power levels to obtain image processing results under different power levels, and the test results under different power levels are used to determine whether the surface-mount LED under test has passed the test.
[0009] Optionally, acquiring the RGB image and light intensity image of the detection station under different power levels includes:
[0010] When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the shooting area, and the shooting area is photographed by an industrial camera to obtain the RGB image of the LED under test under different energizing levels.
[0011] When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the scanning area, and the scanning area is scanned by a photometer to obtain the photointensity map of the LED under test under different energizing levels.
[0012] Optionally, aligning the RGB image and the intensity image at each power level in terms of resolution to obtain a group of images to be processed at different power levels includes:
[0013] Obtain the first resolution of the RGB image and the second resolution of the light intensity image under the same power level;
[0014] The light intensity map is sampled using the first resolution as the sampling target to obtain a sampled intensity map, wherein the sampled intensity map has the first resolution;
[0015] Align the RGB image with the sampled intensity image to obtain a group of images to be processed under different power levels.
[0016] Optionally, the step of performing image processing on the image group to be processed under different power levels to obtain image processing results under different power levels includes:
[0017] The aligned RGB images in the image group to be processed under different power levels are converted into HSV images, and the HSV images include hue channels, saturation channels and luminance channels;
[0018] Replace the hue channel in the HSV image with the aligned light intensity map in the image group to be processed under different power levels to obtain the image to be processed under different power levels. The image to be processed includes a luminance channel, a saturation channel and a light intensity channel corresponding to the aligned light intensity map.
[0019] Image processing is performed on the images to be processed under different power levels to obtain image processing results under different power levels.
[0020] Optionally, the step of performing image processing on the image to be processed under different power levels to obtain image processing results under different power levels includes:
[0021] Obtain grayscale histograms of the image to be processed under different power levels, wherein the grayscale histograms include the distribution of grayscale values corresponding to each pixel in the image to be processed;
[0022] At least one target gray value distribution interval is determined in the gray value histogram under different power levels, and each target gray value distribution interval contains at least one target pixel.
[0023] The target pixels in the target gray value distribution range are clustered by location to obtain clustering results at different locations;
[0024] The clustering results at different locations under different power levels are determined as the image processing results of the image to be processed under the corresponding power level.
[0025] Optionally, determining whether the LED under test passes the test based on the image processing results under different power levels includes:
[0026] Under the same power level, the number of different locations is determined based on the clustering results of the different locations;
[0027] If the number of different positions under any power level is different from the number of the surface-mount LEDs under test, then the test result of the surface-mount LEDs under test is determined to be unsuccessful.
[0028] If the number of different positions under all power levels is the same as the number of the LEDs under test, then the pixel distribution of each position under each power level is determined based on the clustering results of the different positions under each power level.
[0029] The test result is determined based on the pixel distribution at each position under each power level.
[0030] Optionally, determining whether the LED under test passes the test based on the pixel distribution at each position under each power level includes:
[0031] Under the current power level, the number and range of pixels at each location are determined based on the pixel distribution at each location;
[0032] If the number and range of pixels at all locations meet the preset number and range of pixels in the current power level, then the test result of the LED under test in the current power level is determined to be a pass test.
[0033] If the test results of the surface-mount LED under test are all passed under all power levels, then the surface-mount LED under test is determined to have passed the test.
[0034] Secondly, embodiments of the present invention also provide a testing system for surface-mount light-emitting diodes (SLEDs), the testing system comprising:
[0035] The power supply module is used to provide forward power to the surface-mount LEDs under test arranged in an array on the testing platform. The testing platform is provided with testing stations for the surface-mount LEDs under test, and the testing stations are provided with forward power supply.
[0036] The image acquisition module is used to acquire the RGB image and light intensity image of the LED under test at different power levels.
[0037] An image alignment module is used to align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and light intensity image under the corresponding power level.
[0038] The image processing module is used to perform image processing on the image group to be processed under different power levels, obtain the image processing results under different power levels, and determine whether the LED under test has passed the test based on the image processing results under different power levels.
[0039] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the testing method for surface-mount light-emitting diodes provided in embodiments of the present invention.
[0040] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the testing method for surface-mount light-emitting diodes provided in the embodiments of the present invention.
[0041] In this embodiment of the invention, a surface-mount LED under test (SUD) arrayed on a testing platform is forward-biased. The testing platform has an array of testing stations for the SUD, and each testing station is equipped with a forward power supply. Under different power levels, the RGB image and light intensity image of the SUD are acquired. The RGB image and light intensity image under each power level are aligned in resolution dimension to obtain image sets to be processed under different power levels. Each image set includes the aligned RGB image and light intensity image under the corresponding power level. Image processing is performed on the image sets under different power levels to obtain image processing results under different power levels. Based on the image processing results under different power levels, it is determined whether the SUD passes the test. The RGB and light intensity images of the surface-mount LEDs under test, arrayed on the testing stage, are acquired under different power levels. These images are then aligned, and image processing is performed on the aligned images to obtain the results for each power level. The success of the test is determined based on these results, eliminating the need for manual observation of the LEDs' emission. Furthermore, because the LEDs are arrayed on the testing stage with a large emission area, the acquired RGB and light intensity images are less affected by environmental factors, thus improving accuracy. Attached Figure Description
[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0043] Figure 1 This is a flowchart of a testing method for a surface-mount light-emitting diode provided in an embodiment of the present invention;
[0044] Figure 2 This is a schematic diagram of the structure of a test system for surface-mount light-emitting diodes provided in an embodiment of the present invention;
[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0047] like Figure 1 As shown, Figure 1 This is a flowchart illustrating a testing method for a surface-mount light-emitting diode (LED) according to an embodiment of the present invention. The testing method for the surface-mount LED includes the following steps:
[0048] 101. Apply forward power to the LED under test array set on the test stage.
[0049] In this embodiment of the invention, the testing platform is arrayed with testing stations for surface-mount LEDs under test (SMTs), each equipped with a forward power supply. The SMTs under test are sampled from the same batch, and the number of SMTs is the same as the number of testing stations on the testing platform. The SMTs are placed in their respective testing stations, which are then powered on to provide a forward power supply. The array can be rectangular. The testing stations can be configured with different forward power levels for the SMTs under test based on parameters transmitted from a higher-level controller.
[0050] 102. Under different power levels, obtain the RGB image and light intensity image of the surface-mount LED under test.
[0051] In this embodiment of the invention, the above-mentioned power supply level can be set according to the test requirements. The power supply level can be set according to the forward voltage VF of the surface mount light-emitting diode. For example, if the forward voltage VF of the surface mount light-emitting diode is 1.5 to 2.5V, the power supply level can be set to three power supply levels: 1.5V, 2.0V, and 2.5V.
[0052] At different power levels, an image acquisition device can be used to acquire images of the LED under test from above the testing platform to obtain the RGB image of the LED. Alternatively, a light intensity acquisition device can be used to sample the light intensity of the LED under test from above the testing platform to obtain the light intensity image of the LED.
[0053] In one possible embodiment, different acquisition resolutions can be set for different power levels. Higher power levels allow for higher acquisition resolutions of RGB and light intensity images, while lower power levels allow for lower acquisition resolutions. For example, at a power level of 1.5V, a lower acquisition resolution can be set to acquire lower-resolution RGB and light intensity images; at a power level of 2.0V, a higher acquisition resolution can be set to acquire higher-resolution RGB and light intensity images; and at power levels of 2.0V and 2.5V, the highest acquisition resolution can be set to acquire the highest-resolution RGB and light intensity images.
[0054] The above-mentioned testing method for surface-mount LEDs can be applied to electronic devices such as servers or host computers. These electronic devices can obtain the corresponding RGB image and light intensity image through wired or wireless communication.
[0055] 103. Align the RGB image and light intensity image under each power level in terms of resolution to obtain the image group to be processed under different power levels.
[0056] In this embodiment of the invention, the image group to be processed includes an RGB image and a light intensity image aligned to the corresponding power level.
[0057] Specifically, for different power levels, since the RGB image and the light intensity image are acquired using different acquisition devices, the acquisition resolution is different. Therefore, the RGB image and the light intensity image can be processed for resolution before alignment so that the RGB image and the light intensity image have the same resolution, and the pixels in the RGB image and the light intensity image can be mapped one-to-one.
[0058] The aligned RGB image is associated with the aligned light intensity image to obtain a group of images to be processed. Each power level corresponds to a group of images to be processed.
[0059] 104. Perform image processing on the image groups to be processed under different power levels to obtain the image processing results under different power levels, and determine whether the surface-mount LED under test has passed the test based on the image processing results under different power levels.
[0060] In this embodiment of the invention, image processing can be performed on the image groups to be processed under different power levels to obtain the image processing results for each power level. The image processing described above can be based on convolutional neural networks or on pixels.
[0061] In convolutional neural network-based image processing, a convolutional neural network-based image detection model can be trained. Supervised training of the model using a dataset yields a trained image detection model. The model then processes a set of images to be processed, outputting the processed image results. These results include detected defect points, each corresponding to a specific location of a light-emitting diode (LED) patch. The dataset includes sample images and labeling. The sample images consist of aligned RGB images and aligned intensity images. The labeling represents the identified defect points in the RGB images. The image detection model outputs an RGB image containing the defect point detection results. If the image detection model outputs an RGB image with defective points, it indicates that the LED under test at the current power level has failed the test. Furthermore, the LED under test with quality problems can be identified from the RGB image with defective points. If the image detection model outputs an RGB image without defective points, it indicates that the LED under test at the current power level has passed the test. If the LED under test at each power level has passed the test, it indicates that the LED under test has passed the test.
[0062] In pixel-based image processing, the brightest pixel in the RGB image and the most intense pixel in the intensity image can be calculated. If the positions of the brightest and most intense pixels coincide, and the number of these pixels equals the number of LEDs under test, then the LED under test passes the test at the current power level. If the LEDs under test pass the test at every power level, then the LED under test fails the test. If the positions of the brightest and most intense pixels do not coincide, or the number of these pixels does not equal the number of LEDs under test, then the LED under test fails the test at the current power level, and consequently, the LED under test fails the test.
[0063] In this embodiment of the invention, a surface-mount LED under test (SUD) arrayed on a testing stage is forward-biased. The testing stage has an array of testing stations for the SUD, and the testing stations are equipped with a forward power supply. Under different power levels, the RGB image and light intensity image of the SUD are acquired. The RGB image and light intensity image under each power level are aligned in resolution dimension to obtain image sets to be processed under different power levels. Each image set includes the aligned RGB image and light intensity image under the corresponding power level. Image processing is performed on the image sets under different power levels to obtain image processing results under different power levels. Based on the image processing results under different power levels, it is determined whether the SUD passes the test. The RGB and light intensity images of the surface-mount LEDs under test, arrayed on the testing stage, are acquired under different power levels. These images are then aligned, and image processing is performed on the aligned images to obtain the results for each power level. The success of the test is determined based on these results, eliminating the need for manual observation of the LEDs' emission. Furthermore, because the LEDs are arrayed on the testing stage with a large emission area, the acquired RGB and light intensity images are less affected by environmental factors, thus improving accuracy.
[0064] Optionally, in the step of acquiring the RGB image and light intensity image of the testing station under different power levels, when the forward power-on time reaches a preset time under different power levels, the array area of the LED under test is used as the shooting area, and an industrial camera is used to capture the shooting area to obtain the RGB image of the LED under test under different power levels; when the forward power-on time reaches a preset time under different power levels, the array area of the LED under test is used as the scanning area, and a light intensity meter is used to scan the scanning area to obtain the light intensity image of the LED under test under different power levels.
[0065] In this embodiment of the invention, the preset time is the time for the power-on to stabilize, such as 5 seconds or 10 seconds after power-on. Different power-on levels can have different preset times. The higher the power-on level, the longer the preset time. When the forward power-on time reaches the preset time, the RGB image and light intensity image of the LED under test are collected, which can make the collected RGB image and light intensity image more stable.
[0066] Specifically, the image acquisition device is an industrial camera, which is positioned above the testing platform and directly facing the LED under test on the testing platform. By adjusting the shooting parameters of the industrial camera, the industrial camera can be aimed at the array area where the LED under test is located. This array area is the shooting area of the industrial camera. When the forward energization time reaches a preset time under different energization levels, the industrial camera takes pictures of the shooting area to obtain the RGB images of the LED under test under different energization levels.
[0067] The aforementioned light intensity acquisition device can be a movable light intensity meter positioned above the testing platform and facing the testing platform. The array area of the LED under test is used as the scanning area. When the forward energization time reaches a preset time, the light intensity of the scannable area is acquired by moving the light intensity meter to obtain the light intensity map of the LED under test under different energization levels.
[0068] Optionally, in the step of aligning the RGB image and intensity image under each power level in the dimension of resolution to obtain the image group to be processed under different power levels, the first resolution of the RGB image and the second resolution of the intensity image under the same power level can be obtained; the intensity image is sampled with the first resolution as the sampling target to obtain the sampled intensity image, which has the first resolution; the RGB image and the sampled intensity image are aligned to obtain the image group to be processed under different power levels.
[0069] In this embodiment of the invention, generally speaking, the resolution of the industrial camera is higher than that of the moving light intensity meter; that is, the first resolution of the RGB image is greater than the second resolution of the light intensity image. In this embodiment, the first resolution of the RGB image and the second resolution of the light intensity image differ under different power levels. Increasing the power level can increase both the first resolution of the RGB image and the second resolution of the light intensity image. Specifically, increasing the first resolution of the RGB image can be achieved by adjusting the resolution of the industrial camera, while increasing the second resolution of the light intensity image can be achieved by adjusting the moving speed and scanning frame rate of the light intensity meter.
[0070] For RGB and intensity maps of the same resolution, the intensity map can be scaled to match the resolution of the RGB map. With the RGB and intensity maps at the same resolution, they can be aligned so that pixels in the RGB map correspond one-to-one with pixels in the intensity map. Alternatively, linear interpolation can be performed on the intensity map to increase its resolution to match that of the RGB map.
[0071] Aligning the RGB image with the intensity image in terms of resolution facilitates subsequent image processing.
[0072] Optionally, in the step of performing image processing on the image group to be processed under different power levels to obtain image processing results under different power levels, the aligned RGB images in the image group to be processed under different power levels can be converted into HSV images, the HSV images including hue channels, saturation channels, and luminance channels; the aligned light intensity images in the image group to be processed under different power levels replace the hue channels in the HSV images to obtain images to be processed under different power levels, the images to be processed including luminance channels, saturation channels, and light intensity channels corresponding to the aligned light intensity images; and image processing is performed on the images to be processed under different power levels to obtain image processing results under different power levels.
[0073] In this embodiment of the invention, an RGB-to-HSV conversion tool or algorithm can be used to convert the aligned RGB images in the image group to HSV images. For each power level, the aligned RGB images need to be converted to HSV images. The converted HSV images have the same resolution as the RGB images.
[0074] The HSV diagram above includes three channels: hue channel, saturation channel, and luminance channel. Considering that the LEDs under test are from the same batch and belong to the same color category, the hue channel in the HSV diagram can be ignored.
[0075] Since the intensity map is a single-channel image and has the same resolution as the RGB image, it also has the same resolution as the HSV image. Therefore, the intensity map can replace the tone channel in the HSV image to obtain the image to be processed. This replacement operation is performed for each power level, thus each power level has a corresponding image to be processed.
[0076] In this embodiment of the invention, after converting the RGB image to an HSV image, the hue channel in the HSV image is replaced with a light intensity image. This removes invalid hue information, thereby increasing the information richness of the image to be processed while reducing the amount of data in subsequent image processing.
[0077] After obtaining the image to be processed, image processing can be performed on it to obtain the corresponding image processing result. The above image processing can be based on convolutional neural networks or pixel-based image processing.
[0078] In convolutional neural network-based image processing, a convolutional neural network-based image detection model can be trained. This model is then trained in a supervised manner using a dataset. The image to be processed is input into the trained model, and the model outputs the processed image, which includes detected defect points. Each defect point corresponds to a specific location of a light-emitting diode (LED) patch under test. The dataset includes sample images and sample labels. The sample images are obtained by replacing the tone channels of the sample HSV image with sample intensity images. The sample labels are the marked defect points in the sample images. The output of the image detection model is an output image with the defect point detection results, and this output image has the same resolution as the sample images. If the image detection model outputs an image with defective points, it indicates that the LED under test at the current power level has failed the test. Furthermore, the LED with quality problems can be identified from the output image with defective points. If the image detection model outputs an image without defective points, it indicates that the LED under test at the current power level has passed the test. If the LED under test at each power level has passed the test, it indicates that the LED under test has passed the test.
[0079] In pixel-based image processing, the pixels with local maxima in the image can be calculated. If the number of pixels with local maxima equals the number of LEDs under test, and the positions of the pixels with local maxima coincide with the positions of the LEDs under test, then the LEDs under test pass the test at the current power level. If the LEDs under test pass the test at every power level, then the LEDs under test pass the test. If the positions of the pixels with local maxima do not coincide with the positions of the LEDs under test, or if the number of pixels with local maxima does not equal the number of LEDs under test, then the LEDs under test fail the test at the current power level, and therefore, the LEDs under test fail the test.
[0080] Optionally, in the step of processing images under different power levels to obtain image processing results under different power levels, gray-level histograms of the images to be processed under different power levels can be obtained. The gray-level histograms include the distribution of gray values corresponding to each pixel in the image to be processed. At least one target gray-level value distribution interval is determined from the gray-level histograms under different power levels, and each target gray-level value distribution interval contains at least one target pixel. The target pixels in the target gray-level value distribution intervals are clustered at their locations to obtain clustering results at different locations. The clustering results at different locations under different power levels are determined as the image processing results of the image to be processed under the corresponding power level.
[0081] In this embodiment of the invention, the image to be processed can be converted to grayscale to obtain a grayscale image of the image to be processed. Histogram statistics are then performed on the grayscale image of the image to be processed to obtain a grayscale histogram of the image to be processed. For each power level, there is a corresponding grayscale histogram. The aforementioned grayscale histogram can represent the distribution range of grayscale values. The N largest grayscale value distribution ranges can be determined as the target grayscale value distribution area, where N is greater than or equal to 1. N is related to the granularity of the grayscale value distribution range division; the smaller the granularity, the larger N is, and vice versa, with N being at least 1.
[0082] After determining the target grayscale value distribution range, the pixels within that range can be identified as target pixels. These target pixels can be clustered by location, grouping pixels with similar locations into a single cluster. Each cluster can contain at least one target pixel. Each clustering result corresponds to a specific location.
[0083] It is understandable that if the LEDs under test are of good quality, then each cluster should correspond to the location of one LED under test, meaning the number of clusters should be the same as the number of LEDs under test. The clustering results are then used as the image processing results for the image to be processed. These image processing results can include the number of clusters, the center position of each cluster, etc.
[0084] Optionally, in the step of determining whether the LED under test has passed the test based on the image processing results under different power levels, the number of LEDs at different locations can be determined based on the clustering results at different locations under the same power level. If the number of LEDs at different locations under any power level is different from the number of LEDs under test, then the test result of the LED under test is determined to have failed the test. If the number of LEDs at different locations under all power levels is the same as the number of LEDs under test, then the pixel distribution at each location under each power level is determined based on the clustering results at different locations under each power level. The test result of the LED under test is then determined based on the pixel distribution at each location under each power level.
[0085] In this embodiment of the invention, the above image processing result is a clustering result of different locations. For any power level, each clustering result corresponds to a location, and the number of clustering results corresponds to the number of different locations. If the number of different locations is not the same as the number of LEDs under test, it means that there are more light-emitting points in the image to be processed than the number of LEDs under test, or fewer light-emitting points in the image to be processed than the number of LEDs under test. Both are abnormal situations. At this time, it can be directly determined that the test result of the LED under test has failed the test.
[0086] If the number of LEDs at different locations under all power levels is the same as the number of LEDs under test, it indicates that there are emitting points in the image to be processed that equal the number of LEDs under test. In this case, it is necessary to further determine whether there are any abnormal emitting points. The pixel distribution of each cluster result can be determined; based on the pixel distribution of each cluster result, it can be further determined whether the LEDs under test have passed the test.
[0087] Furthermore, the image processing results may include the number of clusters and the center position of the clusters. Each cluster may include at least one target pixel. If the number of clusters is not the same as the number of LEDs under test, it indicates that there are more light-emitting points in the image being processed than the number of LEDs under test, or fewer light-emitting points in the image being processed than the number of LEDs under test. Both of these are abnormal situations, and in this case, it can be directly determined that the test result of the LEDs under test has failed the test.
[0088] If the number of clusters is the same as the number of LEDs under test, it indicates that there are light-emitting points in the image equal to the number of LEDs under test. In this case, it is necessary to further determine whether there are any abnormal light-emitting points. The overlap between the center position of each cluster and the corresponding LED under test can be determined. Based on the overlap, it can be further determined whether the LED under test has passed the test. If the positions overlap, it means the LED under test has passed the test at the corresponding power level. If the LED under test has passed the test at all power levels, then the LED under test has passed the test. If the positions do not overlap, it means the LED under test has failed the test at the corresponding power level.
[0089] Optionally, in the step of determining whether the surface-mount LED under test passes the test based on the pixel distribution at each position under each power level, the number and range of pixels at each position can be determined based on the pixel distribution at each position under the current power level. If the number and range of pixels at all positions meet the preset number and range of pixels in the current power level, then the test result of the surface-mount LED under test under the current power level is determined to be a pass. If the test results of the surface-mount LED under test under all power levels are all pass, then the surface-mount LED under test is determined to be a pass.
[0090] In this embodiment of the invention, for the current power level, the number of pixels at each location can be the number of target pixels in the corresponding clustering result, and the distribution range can be the location of the target pixel farthest from the cluster center in the corresponding clustering result. It should be noted that since the LED under test acts as a light source, its irradiance diffuses from the light source. Therefore, the number and distribution range of pixels at a certain location can be used to determine whether that location matches the pixel distribution number and range of a light source. If the pixel distribution number and distribution range at a location match the pixel distribution number and distribution range of a light source, then that location belongs to a light source. If the light source location coincides with the LED under test, then the corresponding LED under test is of good quality and the test passes. If the light source location does not coincide with the LED under test, and the number of pixels at different locations is the same as the number of LEDs under test, then there is an extra light source that is not the LED under test, indicating a quality problem in this batch of LEDs under test, and the test fails. If the number and range of pixels at a location do not match the number and range of pixels at the light source, then that location is not a light source. If the location of the light source coincides with the LED under test, then the LED under test is not emitting light normally, and there is a quality problem with the LED under test. If the location of the light source does not coincide with the LED under test, then there is a quality problem with the batch of LEDs under test, and the test fails.
[0091] If the test results of the surface-mount LED under test are all passed under all power levels, then the surface-mount LED under test is determined to have passed the test.
[0092] like Figure 2 As shown, this embodiment of the invention provides a testing system for surface-mount light-emitting diodes (SLEDs). The testing system for SLEDs includes:
[0093] The power supply module 201 is used to provide forward power to the surface-mount LEDs under test arranged in an array on the testing platform. The testing platform is provided with a testing station for the surface-mount LEDs under test, and the testing station is provided with a forward power supply.
[0094] Image acquisition module 202 is used to acquire the RGB image and light intensity image of the LED under test at different power levels.
[0095] Image alignment module 203 is used to align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and light intensity image under the corresponding power level.
[0096] The image processing module 204 is used to perform image processing on the image group to be processed under different power levels, obtain the image processing results under different power levels, and determine whether the LED under test has passed the test based on the image processing results under different power levels.
[0097] Optionally, the image acquisition module 202 is further configured to, when the forward energizing time reaches a preset time under different energizing levels, take the array area of the LED under test as the shooting area and take pictures of the shooting area with an industrial camera to obtain RGB images of the LED under test under different energizing levels; when the forward energizing time reaches a preset time under different energizing levels, take the array area of the LED under test as the scanning area and take the scanning area with a light intensity meter to obtain light intensity images of the LED under test under different energizing levels.
[0098] Optionally, the image alignment module 203 is further configured to obtain the first resolution of the RGB image and the second resolution of the light intensity image under the same power level; to sample the light intensity image with the first resolution as the sampling target to obtain a sampled intensity image, wherein the sampled intensity image has the first resolution; and to align the RGB image with the sampled intensity image to obtain a group of images to be processed under different power levels.
[0099] Optionally, the image processing module 204 is further configured to convert the aligned RGB images in the image group to be processed under different power levels into HSV images, wherein the HSV images include a hue channel, a saturation channel, and a luminance channel; replace the hue channel in the HSV images with the aligned light intensity images in the image group to be processed under different power levels to obtain images to be processed under different power levels, wherein the images to be processed include a luminance channel, a saturation channel, and a light intensity channel corresponding to the aligned light intensity images; and perform image processing on the images to be processed under different power levels to obtain image processing results under different power levels.
[0100] Optionally, the image processing module 204 is further configured to acquire grayscale histograms of the image to be processed under different power levels, wherein the grayscale histograms include the grayscale value distribution corresponding to each pixel in the image to be processed; determine at least one target grayscale value distribution interval in the grayscale histograms under different power levels, wherein each target grayscale value distribution interval contains at least one target pixel; perform position clustering on the target pixels in the target grayscale value distribution intervals to obtain clustering results at different positions; and determine the clustering results at different positions under different power levels as the image processing results of the image to be processed under the corresponding power level.
[0101] Optionally, the image processing module 204 is further configured to determine the number of different positions based on the clustering results of the different positions under the same power level; if the number of different positions under any power level is different from the number of the LEDs under test, then the test result of the LED under test is determined to have failed the test; if the number of different positions under all power levels is the same as the number of LEDs under test, then the pixel distribution of each position under each power level is determined based on the clustering results of the different positions under each power level; and the test result of the LED under test is determined based on the pixel distribution of each position under each power level.
[0102] Optionally, the image processing module 204 is further configured to determine the number and range of pixels at each location based on the pixel distribution at each location under the current power level; if the number and range of pixels at all locations conform to the preset number and range of pixels in the current power level, then the test result of the LED under test under the current power level is determined to be a pass; if the test results of the LED under test under all power levels are all pass, then the LED under test is determined to be a pass.
[0103] The testing system for surface-mount LEDs provided in this embodiment of the invention can realize all the processes implemented by the testing method for surface-mount LEDs in the above-described method embodiments, and can achieve the same beneficial effects. To avoid repetition, further details are omitted here.
[0104] See Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program for testing surface-mount light-emitting diodes stored in the memory 302 and executable on the processor 301, wherein:
[0105] The processor 301 is used to call the computer program stored in the memory 302 and perform the following steps:
[0106] A forward-biased LED array is arranged on a testing platform, wherein the testing platform is provided with a testing station for the LED array and the testing station is provided with a forward power supply.
[0107] Under different power levels, the RGB image and light intensity image of the LED under test were obtained respectively.
[0108] Align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and light intensity image under the corresponding power level.
[0109] Image processing is performed on the image group to be processed under different power levels to obtain image processing results under different power levels, and the test results under different power levels are used to determine whether the surface-mount LED under test has passed the test.
[0110] Optionally, the processor 301 performs the following steps: acquiring the RGB image and light intensity image of the detection station under different power levels, including:
[0111] When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the shooting area, and the shooting area is photographed by an industrial camera to obtain the RGB image of the LED under test under different energizing levels.
[0112] When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the scanning area, and the scanning area is scanned by a photometer to obtain the photointensity map of the LED under test under different energizing levels.
[0113] Optionally, the step of processor 301 aligning the RGB image and the intensity image at each power level in the resolution dimension to obtain a group of images to be processed at different power levels includes:
[0114] Obtain the first resolution of the RGB image and the second resolution of the light intensity image under the same power level;
[0115] The light intensity map is sampled using the first resolution as the sampling target to obtain a sampled intensity map, wherein the sampled intensity map has the first resolution;
[0116] Align the RGB image with the sampled intensity image to obtain a group of images to be processed under different power levels.
[0117] Optionally, the image processing performed by the processor 301 on the image group to be processed under different power levels to obtain image processing results under different power levels includes:
[0118] The aligned RGB images in the image group to be processed under different power levels are converted into HSV images, and the HSV images include hue channels, saturation channels and luminance channels;
[0119] Replace the hue channel in the HSV image with the aligned light intensity map in the image group to be processed under different power levels to obtain the image to be processed under different power levels. The image to be processed includes a luminance channel, a saturation channel and a light intensity channel corresponding to the aligned light intensity map.
[0120] Image processing is performed on the images to be processed under different power levels to obtain image processing results under different power levels.
[0121] Optionally, the image processing performed by the processor 301 on the image to be processed under different power levels to obtain image processing results under different power levels includes:
[0122] Obtain grayscale histograms of the image to be processed under different power levels, wherein the grayscale histograms include the distribution of grayscale values corresponding to each pixel in the image to be processed;
[0123] At least one target gray value distribution interval is determined in the gray value histogram under different power levels, and each target gray value distribution interval contains at least one target pixel.
[0124] The target pixels in the target gray value distribution range are clustered by location to obtain clustering results at different locations;
[0125] The clustering results at different locations under different power levels are determined as the image processing results of the image to be processed under the corresponding power level.
[0126] Optionally, the process executed by processor 301 to determine whether the LED under test passes the test based on the image processing results under different power levels includes:
[0127] Under the same power level, the number of different locations is determined based on the clustering results of the different locations;
[0128] If the number of different positions under any power level is different from the number of the surface-mount LEDs under test, then the test result of the surface-mount LEDs under test is determined to be unsuccessful.
[0129] If the number of different positions under all power levels is the same as the number of the LEDs under test, then the pixel distribution of each position under each power level is determined based on the clustering results of the different positions under each power level.
[0130] The test result is determined based on the pixel distribution at each position under each power level.
[0131] Optionally, the processor 301's execution of determining whether the LED under test passes the test based on the pixel distribution at each position under each power level includes:
[0132] Under the current power level, the number and range of pixels at each location are determined based on the pixel distribution at each location;
[0133] If the number and range of pixels at all locations meet the preset number and range of pixels in the current power level, then the test result of the LED under test in the current power level is determined to be a pass test.
[0134] If the test results of the surface-mount LED under test are all passed under all power levels, then the surface-mount LED under test is determined to have passed the test.
[0135] The electronic device provided in this embodiment of the invention can implement all the processes of the surface-mount LED testing method in the above-described method embodiments, and can achieve the same beneficial effects. To avoid repetition, further details are omitted here.
[0136] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the surface-mount light-emitting diode testing method provided in this invention and achieves the same technical effect. To avoid repetition, it will not be described again here.
[0137] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.
[0138] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.
Claims
1. A testing method for surface-mount light-emitting diodes, characterized in that, include: A forward-biased LED array is arranged on a testing platform, wherein the testing platform is provided with a testing station for the LED array and the testing station is provided with a forward power supply. Under different power levels, the RGB image and light intensity image of the LED under test were obtained respectively. Align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and the light intensity image under the corresponding power level. Image processing is performed on the image group to be processed under different power levels to obtain image processing results under different power levels, and the test results under different power levels are used to determine whether the patch LED under test has passed the test. The image processing of the image group under different power levels to obtain image processing results under different power levels includes: The aligned RGB images in the image group to be processed under different power levels are converted into HSV images, and the HSV images include hue channels, saturation channels and luminance channels; Replace the hue channel in the HSV image with the aligned light intensity map in the image group to be processed under different power levels to obtain the image to be processed under different power levels. The image to be processed includes a luminance channel, a saturation channel and a light intensity channel corresponding to the aligned light intensity map. Image processing is performed on the images to be processed under different power levels to obtain image processing results under different power levels.
2. The method as described in claim 1, characterized in that, The acquisition of the RGB image and light intensity image of the detection station under different power levels includes: When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the shooting area, and the shooting area is photographed by an industrial camera to obtain the RGB image of the LED under test under different energizing levels. When the forward energizing time reaches a preset time under different energizing levels, the array area of the LED under test is used as the scanning area, and the scanning area is scanned by a photometer to obtain the photointensity map of the LED under test under different energizing levels.
3. The method as described in claim 2, characterized in that, The step of aligning the RGB image and the intensity image at each power level in terms of resolution to obtain a group of images to be processed at different power levels includes: Obtain the first resolution of the RGB image and the second resolution of the light intensity image under the same power level; The light intensity map is sampled using the first resolution as the sampling target to obtain a sampled intensity map, wherein the sampled intensity map has the first resolution; Align the RGB image with the sampled intensity image to obtain a group of images to be processed under different power levels.
4. The method as described in claim 1, characterized in that, The step of processing the image to be processed under different power levels to obtain image processing results under different power levels includes: Obtain grayscale histograms of the image to be processed under different power levels, wherein the grayscale histograms include the distribution of grayscale values corresponding to each pixel in the image to be processed; At least one target gray value distribution interval is determined in the gray value histogram under different power levels, and each target gray value distribution interval contains at least one target pixel. The target pixels in the target gray value distribution range are clustered by location to obtain clustering results at different locations; The clustering results at different locations under different power levels are determined as the image processing results of the image to be processed under the corresponding power level.
5. The method as described in claim 4, characterized in that, The step of determining whether the LED under test passes the test based on the image processing results under different power levels includes: Under the same power level, the number of different locations is determined based on the clustering results of the different locations; If the number of different positions under any power level is different from the number of the surface-mount LEDs under test, then the test result of the surface-mount LEDs under test is determined to be unsuccessful. If the number of different positions under all power levels is the same as the number of the LEDs under test, then the pixel distribution of each position under each power level is determined based on the clustering results of the different positions under each power level. The test result is determined based on the pixel distribution at each position under each power level.
6. The method as described in claim 5, characterized in that, The process of determining whether the LED under test passes the test based on the pixel distribution at each position under various power levels includes: Under the current power level, the number and range of pixels at each location are determined based on the pixel distribution at each location; If the number and range of pixels at all locations meet the preset number and range of pixels in the current power level, then the test result of the LED under test in the current power level is determined to be a pass test. If the test results of the surface-mount LED under test are all passed under all power levels, then the surface-mount LED under test is determined to have passed the test.
7. A testing system for surface-mount light-emitting diodes, characterized in that, The testing system for the surface-mount LED includes: The power supply module is used to provide forward power to the surface-mount LEDs under test arranged in an array on the testing platform. The testing platform is provided with testing stations for the surface-mount LEDs under test, and the testing stations are provided with forward power supply. The image acquisition module is used to acquire the RGB image and light intensity image of the LED under test at different power levels. An image alignment module is used to align the RGB image and the light intensity image under each power level in the dimension of resolution to obtain a group of images to be processed under different power levels. The group of images to be processed includes the aligned RGB image and light intensity image under the corresponding power level. The image processing module is used to perform image processing on the image group to be processed under different power levels, obtain the image processing results under different power levels, and determine whether the LED under test passes the test based on the image processing results under different power levels. The image processing of the image group under different power levels to obtain image processing results under different power levels includes: The aligned RGB images in the image group to be processed under different power levels are converted into HSV images, and the HSV images include hue channels, saturation channels and luminance channels; Replace the hue channel in the HSV image with the aligned light intensity map in the image group to be processed under different power levels to obtain the image to be processed under different power levels. The image to be processed includes a luminance channel, a saturation channel and a light intensity channel corresponding to the aligned light intensity map. Image processing is performed on the images to be processed under different power levels to obtain image processing results under different power levels.
8. An electronic device, characterized in that, include: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the test method for surface-mount light-emitting diodes as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the test method for surface-mount light-emitting diodes as described in any one of claims 1 to 6.