A method for detecting defects in a gallium oxide-based light emitting diode epitaxial structure

Abstract

The application discloses a gallium oxide-based light-emitting diode epitaxial structure defect detection method and relates to the electrical field, and comprises the following steps: S1, dividing a target epitaxial structure region into a plurality of sub-regions, obtaining color abnormal values and area abnormal area ratios of each sub-region under different spectrums, and forming static detection data; S2, performing static defect detection based on the static detection data, and if there is no static defect, performing dynamic detection and obtaining a peak luminous intensity deviation; and S3, publishing a static or dynamic defect early warning according to dynamic detection data. Through the multi-spectrum collaborative analysis and the dynamic and static combined detection, the application significantly improves the comprehensiveness and accuracy of defect detection.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and in particular to a method for detecting defects in the epitaxial structure of gallium oxide-based light-emitting diodes. Background Technology

[0002] In the field of semiconductor technology, gallium oxide (GaO)-based light-emitting diodes (LEDs) have attracted much attention due to their excellent photoelectric performance and broad application prospects. As the core component of GaO-based LEDs, the quality of the epitaxial structure directly affects the device's luminous efficiency, stability, and lifespan. Therefore, defect detection methods for the epitaxial structure of GaO-based LEDs have become a crucial aspect of technological development. However, existing epitaxial structure defect detection methods have significant shortcomings in practical applications, making it difficult to meet the demands for high-precision and comprehensive detection.

[0003] Existing technologies typically cannot perform static defect detection on sub-regions of gallium oxide-based light-emitting diode (LED) epitaxial structures under spectral illumination, resulting in inaccurate static defect detection results. For example, traditional methods often rely on macroscopic observation or single-spectral analysis, making it difficult to capture micrometer-level color anomalies and deep structural defects, thus missing some critical defect information. Furthermore, existing detection methods are usually limited to a single dimension of either static or dynamic detection, failing to simultaneously detect static and dynamic defects in epitaxial structure sub-regions, leading to a lack of comprehensiveness in the detection process. This limitation restricts defect tracing and process optimization, impacting overall detection efficiency and reliability.

[0004] To address the aforementioned issues, a defect detection method capable of integrating multiple detection techniques and achieving high-precision, multi-dimensional detection is urgently needed. This invention aims to improve the accuracy and comprehensiveness of defect detection in the epitaxial structure of gallium oxide-based light-emitting diodes (LEDs) by introducing multispectral synergistic detection, pixel-level quantization analysis, and a combined dynamic and static detection strategy, thereby providing technical support for improving device performance and manufacturing processes. Summary of the Invention

[0005] The purpose of this invention is to provide a method for detecting defects in the epitaxial structure of gallium oxide-based light-emitting diodes (LEDs). This aims to improve the comprehensiveness and accuracy of the defect detection process.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode, comprising the following specific steps:

[0007] S1: Divide the target epitaxial structure region into several epitaxial structure sub-regions, obtain the color anomaly value of each epitaxial structure sub-region under different types of detection spectra, and obtain the area ratio of the region anomaly corresponding to each epitaxial structure sub-region based on the color anomaly value to obtain the static detection data of the epitaxial structure.

[0008] S2: Perform static structural defect detection on gallium oxide-based light-emitting diodes based on the static detection data of the epitaxial structure. If there are no static structural defects in the gallium oxide-based light-emitting diodes, perform dynamic structural defect detection on the gallium oxide-based light-emitting diodes and obtain the dynamic detection data of the epitaxial structure based on the detection results.

[0009] S3: Provide early warning of structural defects for gallium oxide-based light-emitting diodes based on dynamic detection data of the epitaxial structure.

[0010] Furthermore, step S1 also includes the following specific steps:

[0011] S11: Obtain the epitaxial structure region corresponding to the gallium oxide-based light-emitting diode to obtain the target epitaxial structure region, divide the target epitaxial structure region into several epitaxial structure sub-regions with equal area, and arbitrarily select a sample epitaxial sub-region from the multiple obtained epitaxial structure sub-regions.

[0012] S12: Perform color anomaly analysis on the extended sub-region of the sample, and obtain the area ratio of the abnormal region corresponding to the extended sub-region of the sample based on the analysis results;

[0013] S13: Obtain the area ratio of the abnormal region corresponding to each sub-region of the extensional structure to obtain the static detection data of the extensional structure.

[0014] Furthermore, step S12 also includes the following specific steps:

[0015] S121: In the process of analyzing the color anomaly of the sample's extended sub-region, several different types of detection spectra are set for the sample's extended sub-region, and the set detection spectra are named H1 detection spectrum to Ha detection spectrum respectively.

[0016] S122: Divide the sample extension sub-region into several region pixels, and arbitrarily select one sample region pixel from the multiple obtained region pixels;

[0017] S123: Perform hue analysis on the pixels of the sample area under H1 detection spectrum illumination, and obtain the H1 hue deviation value corresponding to the pixels of the sample area based on the analysis results.

[0018] Furthermore, step S12 also includes the following specific steps:

[0019] S124: Obtain the hue deviation values ​​of the pixels in the sample area under the illumination of H2 detection spectrum to Ha detection spectrum, and obtain the hue deviation values ​​from H1 to Ha.

[0020] S125: The average of the obtained H1 hue deviation values ​​to Ha hue deviation values ​​is calculated to obtain the average hue deviation value corresponding to the pixel points in the sample area.

[0021] S126: Obtain the hue deviation reference range corresponding to the pixel points in the sample area. If the average hue deviation value corresponding to the pixel points in the sample area is within the hue deviation reference range, then mark the pixel points in the sample area as hue normal pixels. If the average hue deviation value corresponding to the pixel points in the sample area is not within the hue deviation reference range, then mark the pixel points in the sample area as hue abnormal pixels.

[0022] Furthermore, step S12 also includes the following specific steps:

[0023] S127: Obtain the average hue deviation value corresponding to each region pixel, and divide the obtained region pixels into normal hue pixels and abnormal hue pixels according to the average hue deviation value.

[0024] S128: Count the number of normal hue pixels in the sample extension sub-region to obtain the number of normal hue pixels. Count the number of abnormal hue pixels in the sample extension sub-region to obtain the number of abnormal hue pixels. Calculate the ratio of the abnormal region area corresponding to the sample extension sub-region by combining the number of normal hue pixels and the number of abnormal hue pixels.

[0025] The area ratio of the abnormal region corresponding to the extended sub-region of the sample is calculated using the following formula:

[0026] ;

[0027] Where Smb is the area ratio of the abnormal region corresponding to the sample extension sub-region, Zcs is the number of normal hue pixels, and Ycs is the number of abnormal hue pixels.

[0028] Furthermore, step S123 also includes the following specific steps:

[0029] The sample area pixels are illuminated using H1 detection spectrum, and the H1 pixel image is obtained by real-time image acquisition of the sample area pixels under H1 detection spectrum illumination using an image acquisition device.

[0030] The H1 pixel image is decomposed using the HSV color model. Based on the decomposition results, the hue, saturation and lightness values ​​of the sample region pixels are obtained respectively.

[0031] The hue reference value, saturation reference value, and lightness reference value of the pixels in the sample area under the illumination of the H1 detection spectrum are obtained respectively.

[0032] The hue value, saturation value, lightness value, hue reference value, saturation reference value, and lightness reference value are used to calculate the H1 hue deviation value corresponding to the pixel point in the sample area.

[0033] The hue deviation value of H1 is calculated using the following formula:

[0034] ;

[0035] Where Sph1 is the H1 hue deviation value corresponding to the pixel point in the sample area, Syy is the hue value, Myy is the saturation value, Byy is the brightness value, Syj is the hue reference value, Myj is the saturation reference value, and Byj is the brightness reference value.

[0036] Furthermore, step S2 also includes the following specific steps:

[0037] S21: Obtain static detection data of the epitaxial structure, obtain the area ratio of the abnormal region corresponding to each sub-region of the epitaxial structure based on the static detection data of the epitaxial structure, and calculate the average value of the obtained multiple abnormal region area ratios to obtain the average value of the abnormal region area ratio.

[0038] S22: Obtain the acceptable range of abnormal area ratio. If the average value of abnormal area ratio is within the acceptable range of abnormal area ratio, it is determined that the corresponding gallium oxide-based light-emitting diode does not have a static structural defect. If the average value of abnormal area ratio is not within the acceptable range of abnormal area ratio, it is determined that the corresponding gallium oxide-based light-emitting diode has a static structural defect, and static structural detection data is obtained.

[0039] S23: Perform dynamic epitaxial structure detection on gallium oxide-based light-emitting diodes without static structural defects, and obtain the peak emission brightness deviation based on the detection results;

[0040] S24: Define the peak emission brightness deviation and static structure detection data as the epitaxial structure dynamic detection data.

[0041] Furthermore, step S23 also includes the following specific steps:

[0042] S231: In the process of dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, several feature detection powers are set, and a sample feature detection power is selected from the multiple feature detection powers obtained.

[0043] S232: Perform light emission detection on a gallium oxide-based light-emitting diode at the sample feature detection power to obtain the light emission brightness deviation corresponding to the sample feature detection power;

[0044] S233: Obtain the luminous brightness deviation corresponding to the detection power of each feature, compare the numerical values ​​of the multiple luminous brightness deviations obtained, and mark the luminous brightness deviation with the largest value as the detection peak luminous brightness deviation.

[0045] Furthermore, step S232 also includes the following specific steps:

[0046] A gallium oxide-based light-emitting diode is connected to a power-on circuit, and the power of the corresponding power-on circuit is adjusted to the feature detection power.

[0047] During the dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, the current time point is marked as the end time point of the cycle, the time point when the gallium oxide-based light-emitting diode starts to emit light is marked as the start time point of the cycle, and the time interval between the start time point and the end time point of the cycle is marked as the luminous intensity analysis cycle.

[0048] The luminous intensity values ​​of gallium oxide-based light-emitting diodes are obtained within the luminous brightness analysis period to obtain the periodic luminous intensity monitoring values.

[0049] Obtain the reference brightness value of the gallium oxide-based light-emitting diode at the sample feature detection power, calculate the difference between the periodic luminous intensity monitoring value and the reference brightness value, and take the absolute value of the obtained difference to obtain the luminous brightness deviation corresponding to the sample feature detection power.

[0050] Furthermore, step S3 also includes the following specific steps:

[0051] S31: Obtain dynamic detection data of the epitaxial structure, and obtain the peak emission brightness deviation and static structure detection data based on the dynamic detection data of the epitaxial structure;

[0052] S32: If a gallium oxide-based light-emitting diode has static structural defects, a static defect warning will be issued directly for the gallium oxide-based light-emitting diode.

[0053] S33: If there are no static structural defects in the gallium oxide-based light-emitting diode, then a dynamic defect warning is given to the gallium oxide-based light-emitting diode based on the detected peak luminous brightness deviation;

[0054] Step S33 further includes the following specific steps:

[0055] If the detected peak luminous brightness deviation falls within the luminous brightness deviation reference range, then there is no need to issue a dynamic defect warning for the gallium oxide-based light-emitting diode.

[0056] If the detected peak luminous intensity deviation is not within the luminous intensity deviation reference range, a dynamic defect warning will be issued for the gallium oxide-based light-emitting diode.

[0057] The beneficial effects of this invention are:

[0058] Through multispectral synergistic analysis and dynamic detection mechanisms, the accuracy and comprehensiveness of defect detection in the epitaxial structure of gallium oxide-based light-emitting diodes have been significantly improved.

[0059] Furthermore, the method achieves precise capture of micron-level color anomalies by quantitatively analyzing the hue deviation values ​​of pixels in the sample area, and has significant advantages, especially in deep structural defects, surface oxide layer uniformity, and doping distribution.

[0060] In particular, by constructing a dynamic benchmark interval and introducing the average hue deviation value as an anomaly detection threshold, the error problem of single threshold judgment is avoided, and the adaptability of the detection standard is improved.

[0061] Furthermore, through the fusion analysis of static and dynamic data, not only was a spatial visualization mapping of defect severity achieved, but also a quantitative basis for subsequent process optimization was provided. Attached Figure Description

[0062] Figure 1 This is a diagram illustrating the implementation steps of the present invention;

[0063] Figure 2 This is a schematic diagram of the sample extensional sub-region in this invention. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0065] Example 1

[0066] Please see Figure 1 This invention provides a technical solution: a method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode, comprising the following specific steps:

[0067] Step S1: Divide the target epitaxial structure region into several epitaxial structure sub-regions, obtain the color anomaly value of each epitaxial structure sub-region under different types of detection spectra, and obtain the area ratio of the region anomaly corresponding to each epitaxial structure sub-region based on the color anomaly value to obtain the static detection data of the epitaxial structure.

[0068] Step S1 further includes the following specific steps:

[0069] Please see Figure 2The epitaxial structure region corresponding to the gallium oxide-based light-emitting diode is obtained to obtain the target epitaxial structure region. The target epitaxial structure region is divided into several epitaxial structure sub-regions with equal area, and a sample epitaxial sub-region is randomly selected from the multiple obtained epitaxial structure sub-regions.

[0070] It should be noted here that:

[0071] In this application, the epitaxial structure region corresponding to the gallium oxide-based light-emitting diode mentioned herein refers to the multilayer physical structure formed by depositing gallium oxide (Ga2O3) and its related compound semiconductor materials layer by layer on a specific substrate through epitaxial growth technology;

[0072] Perform color anomaly analysis on the extended sub-regions of the sample, and obtain the area ratio of the abnormal regions corresponding to the extended sub-regions of the sample based on the analysis results;

[0073] Specifically as follows:

[0074] In the process of analyzing the color anomaly of the sample's epitaxial sub-region, several different types of detection spectra were set for the sample's epitaxial sub-region, and the set detection spectra were named H1 detection spectrum to Ha detection spectrum respectively.

[0075] It should be noted here that:

[0076] In this application, H is the symbol corresponding to the detection spectrum, and a is the quantity value corresponding to the detection spectrum, and a is an integer greater than 0;

[0077] In this application, the H1 detection spectrum may be an infrared spectrum, the H2 detection spectrum may be an ultraviolet spectrum, and the H3 detection spectrum may be a fluorescence spectrum.

[0078] The sample's outer sub-region is divided into several region pixels, and one sample region pixel is randomly selected from the multiple obtained region pixels.

[0079] Perform hue analysis on the pixels in the sample area under H1 detection spectrum illumination, and obtain the H1 hue deviation value corresponding to the pixels in the sample area based on the analysis results;

[0080] Specifically as follows:

[0081] The sample area pixels are illuminated using H1 detection spectrum, and the H1 pixel image is obtained by real-time image acquisition of the sample area pixels under H1 detection spectrum illumination using an image acquisition device.

[0082] The H1 pixel image is decomposed using the HSV color model. Based on the decomposition results, the hue, saturation and lightness values ​​of the sample region pixels are obtained respectively.

[0083] The hue reference value, saturation reference value, and lightness reference value of the pixels in the sample area under the illumination of the H1 detection spectrum are obtained respectively.

[0084] It should be noted here that:

[0085] The hue, saturation, and brightness reference values ​​mentioned here are the optimal hue, saturation, and brightness values ​​of the pixel points in the sample area corresponding to a qualified gallium oxide-based light-emitting diode under H1 detection spectrum illumination.

[0086] The hue value, saturation value, lightness value, hue reference value, saturation reference value, and lightness reference value are used to calculate the H1 hue deviation value corresponding to the pixel point in the sample area.

[0087] The hue deviation value of H1 is calculated using the following formula:

[0088] ;

[0089] Where Sph1 is the H1 hue deviation value corresponding to the pixel point in the sample area, Syy is the hue value, Myy is the saturation value, Byy is the brightness value, Syj is the hue reference value, Myj is the saturation reference value, and Byj is the brightness reference value.

[0090] It should be noted here that:

[0091] In practical applications, if Syy is measured to be 21%, Myy to be 66%, Byy to be 55%, and Syj to be 23%, Myj to be 54%, and Byj to be 71%, then the H1 hue deviation value corresponding to the pixel points in the sample area can be calculated to be 6.3%.

[0092] Repeat the process of obtaining the H1 hue deviation value corresponding to the pixel point in the sample area, and obtain the hue deviation value of the pixel point in the sample area under the illumination of the H2 detection spectrum to the Ha detection spectrum respectively, to obtain the H1 hue deviation value to the Ha hue deviation value.

[0093] The average of the obtained H1 hue deviation values ​​to Ha hue deviation values ​​is calculated to obtain the average hue deviation value corresponding to the pixel points in the sample area.

[0094] Several gallium oxide-based light-emitting diodes (LEDs) with normal structural defect detection results were obtained. The pixel points of the sample area corresponding to each gallium oxide-based LED were obtained to obtain multiple historical sample area pixel points. The average hue deviation value corresponding to each historical sample area pixel point was obtained. The multiple average hue deviation values ​​were compared. The average hue deviation value with the largest value was marked as the first average hue deviation benchmark value, and the average hue deviation value with the smallest value was marked as the second average hue deviation benchmark value. The first average hue deviation benchmark value was set as the upper limit of the interval, and the second average hue deviation benchmark value was set as the lower limit of the interval to obtain the hue deviation benchmark interval corresponding to the pixel points of the sample area.

[0095] If the average hue deviation value corresponding to the pixel in the sample area is within the hue deviation reference range, then the pixel in the sample area is marked as a normal hue pixel.

[0096] If the average hue deviation value corresponding to a pixel in the sample area is not within the hue deviation reference range, then the pixel in the sample area will be marked as a hue aberration pixel.

[0097] It should be noted here that:

[0098] In this application, the hue-normal pixels referred to herein include those whose average hue deviation value is at the boundary of the hue deviation reference range.

[0099] Repeatedly obtain the average hue deviation value corresponding to the pixel points in the sample area, obtain the average hue deviation value corresponding to the pixel points in each area, and divide the obtained area pixel points into normal hue pixel points and abnormal hue pixel points according to the average hue deviation value.

[0100] The number of normal hue pixels in the extended sub-region of the sample is counted to obtain the number of normal hue pixels. The number of abnormal hue pixels in the extended sub-region of the sample is counted to obtain the number of abnormal hue pixels.

[0101] The ratio of the abnormal region area corresponding to the sample's extraterrestrial sub-region is obtained by calculating the number of normal hue pixels and the number of abnormal hue pixels.

[0102] The area ratio of the abnormal region corresponding to the extended sub-region of the sample is calculated using the following formula:

[0103] ;

[0104] Where Smb is the area ratio of the abnormal region corresponding to the sample extension sub-region, Zcs is the number of normal hue pixels, and Ycs is the number of abnormal hue pixels.

[0105] Repeat the ratio of the abnormal region area corresponding to the sample's epitaxial sub-region, and obtain the ratio of the abnormal region area corresponding to each epitaxial sub-region to obtain the static detection data of the epitaxial structure.

[0106] The above step S1 has the following advantages:

[0107] Step S1 achieves high-precision characterization of epitaxial structure defects through multispectral collaborative detection and pixel-level quantization analysis. Its core benefits are reflected in three aspects: First, the target area is discretized into equal-area sub-regions and a pixel color reference database is established. Through the three-dimensional parameter decomposition of hue-saturation-brightness under multispectral (H1 to Ha) illumination, micron-level color anomalies that are difficult to identify by traditional macroscopic detection can be accurately captured. In particular, the infrared spectrum has specific responses to deep structural defects, the ultraviolet spectrum to surface oxide layers, and the fluorescence spectrum to doping uniformity, forming complementary detection dimensions. Second, a dynamic reference interval (the average deviation reference value of the first / second hue) is constructed and the average hue deviation value is introduced as an anomaly judgment threshold. This avoids the absolute error of a single threshold judgment and achieves adaptive calibration of the detection standard through interval setting driven by historical sample data. Finally, through the quantitative conversion of the proportion of abnormal pixels and the area ratio, the color anomaly distribution is transformed into a comparable numerical index. This not only realizes the spatial visualization mapping of the severity of defects but also provides a quantitative improvement basis based on defect density distribution for subsequent process optimization.

[0108] Step S2: Perform static structural defect detection on the gallium oxide-based light-emitting diode based on the static detection data of the epitaxial structure. If there are no static structural defects in the gallium oxide-based light-emitting diode, perform dynamic structural defect detection on the gallium oxide-based light-emitting diode and obtain the dynamic detection data of the epitaxial structure based on the detection results.

[0109] Step S2 further includes the following specific steps:

[0110] Obtain static detection data of the epitaxial structure, obtain the area ratio of the abnormal region corresponding to each sub-region of the epitaxial structure based on the static detection data, and calculate the average value of the obtained multiple abnormal region area ratios to obtain the average value of the abnormal region area ratio.

[0111] Obtain the acceptable range of abnormal area ratio. If the average abnormal area ratio is within the acceptable range, it is determined that the corresponding gallium oxide-based light-emitting diode does not have a static structural defect. If the average abnormal area ratio is not within the acceptable range, it is determined that the corresponding gallium oxide-based light-emitting diode has a static structural defect, and static structural detection data is obtained.

[0112] It should be noted here that:

[0113] In this application, the gallium oxide-based light-emitting diodes involved herein do not have static structural defects, including the case where the average area ratio of the abnormal region is at the boundary of the acceptable range of the abnormal region area ratio.

[0114] It should be noted here that:

[0115] A number of historical gallium oxide-based light-emitting diodes (LEDs) with no abnormalities were obtained from static testing of their epitaxial structures. The average area ratio of the abnormal region corresponding to each historical gallium oxide-based LED was obtained. The average area ratios of the multiple abnormal regions were compared. The average area ratio of the abnormal region with the largest value was marked as the first area ratio benchmark value, and the average area ratio of the abnormal region with the smallest value was marked as the second area ratio benchmark value. The numerical interval formed by the first area ratio benchmark value and the second area ratio benchmark value was marked as the qualified interval of the abnormal region area ratio.

[0116] Dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes without static structural defects is performed, and the deviation of the peak luminous brightness is obtained based on the detection results.

[0117] Specifically as follows:

[0118] In the process of dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, several feature detection powers are set, and a sample feature detection power is selected from the multiple feature detection powers obtained.

[0119] A gallium oxide-based light-emitting diode is connected to a power-on circuit, and the power of the corresponding power-on circuit is adjusted to the feature detection power.

[0120] During the dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, the current time point is marked as the end time point of the cycle, the time point when the gallium oxide-based light-emitting diode starts to emit light is marked as the start time point of the cycle, and the time interval between the start time point and the end time point of the cycle is marked as the luminous intensity analysis cycle.

[0121] The luminous intensity values ​​of gallium oxide-based light-emitting diodes are obtained within the luminous brightness analysis period to obtain the periodic luminous intensity monitoring values.

[0122] It should be noted here that:

[0123] In this application, the luminous intensity referred to herein is specifically the brightness value of a gallium oxide-based light-emitting diode;

[0124] Obtain the reference brightness value of the gallium oxide-based light-emitting diode at the sample feature detection power, calculate the difference between the periodic luminous intensity monitoring value and the reference brightness value, and take the absolute value of the obtained difference to obtain the luminous brightness deviation corresponding to the sample feature detection power.

[0125] Repeat the process of obtaining the luminance deviation corresponding to the sample feature detection power, obtain the luminance deviation corresponding to each feature detection power, compare the numerical values ​​of the obtained multiple luminance deviations, and mark the luminance deviation with the largest value as the detection peak luminance deviation.

[0126] The peak emission brightness deviation and static structure detection data are defined as the epitaxial structure dynamic detection data.

[0127] The following advantages exist in step S2 above:

[0128] Step S2 quickly screens qualified devices and establishes quantitative standards through static detection, and captures power-related defects by combining dynamic detection. Finally, the static and dynamic data are integrated to achieve accurate defect tracing, which significantly improves detection efficiency and defect analysis capabilities.

[0129] Step S3: Perform structural defect early warning for gallium oxide-based light-emitting diodes based on dynamic detection data of the epitaxial structure;

[0130] Step S3 further includes the following specific steps:

[0131] Acquire dynamic detection data of the epitaxial structure, and obtain the peak luminous intensity deviation and static structure detection data based on the dynamic detection data of the epitaxial structure;

[0132] If a gallium oxide-based light-emitting diode has static structural defects, a static defect warning will be issued directly for the gallium oxide-based light-emitting diode.

[0133] If there are no static structural defects in the gallium oxide-based light-emitting diode, dynamic defect warning can be given to the gallium oxide-based light-emitting diode based on the detected peak luminous brightness deviation.

[0134] Specifically as follows:

[0135] Obtain the reference range for luminous brightness deviation. If the detected peak luminous brightness deviation is within the reference range, there is no need to issue a dynamic defect warning for the gallium oxide-based light-emitting diode. If the detected peak luminous brightness deviation is not within the reference range, a dynamic defect warning is issued for the gallium oxide-based light-emitting diode.

[0136] It should be noted here that:

[0137] The reference range for luminous intensity deviation is obtained as follows:

[0138] A number of historical gallium oxide-based light-emitting diodes (LEDs) with no abnormalities in their epitaxial structures were obtained. The peak emission brightness deviation of each historical gallium oxide-based LED was obtained. The values ​​of the obtained peak emission brightness deviations were compared. The peak emission brightness deviation with the largest value was marked as the first brightness deviation reference value, and the peak emission brightness deviation with the smallest value was marked as the second brightness deviation reference value. The numerical range formed by the first brightness deviation reference value and the second brightness deviation reference value was marked as the emission brightness deviation reference range.

[0139] The requirement for issuing dynamic defect warnings for gallium oxide-based light-emitting diodes (LEDs) as described here includes detecting situations where the peak luminous intensity deviation falls within the boundary of the luminous intensity deviation reference range.

[0140] In this application, if a corresponding calculation formula appears, the above calculation formula is a dimensionless calculation. The weighting coefficient, proportional coefficient and other coefficients in the formula are set to quantify each parameter to obtain a result value. The size of the weighting coefficient and proportional coefficient is only required to not affect the proportional relationship between the parameter and the result value.

[0141] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode, characterized in that, Includes the following steps: S1: Divide the target epitaxial structure region into several epitaxial structure sub-regions with equal areas, and obtain color anomaly values ​​for each sub-region under different types of detection spectra. Calculate the area ratio of the region anomaly corresponding to each epitaxial structure sub-region based on the color anomaly values ​​to form static detection data of the epitaxial structure. S11: Divide the epitaxial structure region corresponding to the gallium oxide-based light-emitting diode into several epitaxial structure sub-regions of equal area, and arbitrarily select one sample epitaxial sub-region from the multiple obtained epitaxial structure sub-regions. S12: Perform color anomaly analysis on the extended sub-region of the sample, and obtain the area ratio of the abnormal region corresponding to the extended sub-region of the sample based on the analysis results; S121: In the process of analyzing the color anomaly of the sample's extended sub-region, several different types of detection spectra are set for the sample's extended sub-region, and the set detection spectra are named H1 detection spectrum to Ha detection spectrum respectively. S122: Divide the sample extension sub-region into several region pixels, and arbitrarily select one sample region pixel from the multiple obtained region pixels; S123: Perform hue analysis on the pixels of the sample area under the illumination of the H1 detection spectrum, and obtain the H1 hue deviation value corresponding to the pixels of the sample area based on the analysis results; The sample area pixels are illuminated using H1 detection spectrum, and the H1 pixel image is obtained by real-time image acquisition of the sample area pixels under H1 detection spectrum illumination using an image acquisition device. The H1 pixel image is decomposed using the HSV color model. Based on the decomposition results, the hue, saturation and lightness values ​​of the sample region pixels are obtained respectively. The hue reference value, saturation reference value, and lightness reference value of the pixels in the sample area under the illumination of the H1 detection spectrum are obtained respectively. The hue value, saturation value, lightness value, hue reference value, saturation reference value, and lightness reference value are used to calculate the H1 hue deviation value corresponding to the pixel point in the sample area. The specific formula for calculating the hue deviation value of H1 is as follows: ； Where Sph1 is the H1 hue deviation value corresponding to the pixel point in the sample area, Syy is the hue value, Myy is the saturation value, Byy is the brightness value, Syj is the hue reference value, Myj is the saturation reference value, and Byj is the brightness reference value. S13: Obtain the area ratio of the abnormal region corresponding to each sub-region of the epitaxial structure to obtain the static detection data of the epitaxial structure; S2: Static structural defect detection is performed on gallium oxide-based light-emitting diodes based on epitaxial structure static detection data. If no static structural defects are found in the detection results, dynamic structural defect detection is performed. The peak emission brightness deviation is obtained through dynamic detection, and this deviation is combined with the static detection data to form epitaxial structure dynamic detection data. S3: Based on the dynamic detection data of the epitaxial structure, perform structural defect early warning for gallium oxide-based light-emitting diodes, specifically including two categories: static defect early warning and dynamic defect early warning.

2. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 1, characterized in that, Step S12 further includes the following specific steps: S124: Obtain the hue deviation values ​​of the pixels in the sample area under the illumination of H2 detection spectrum to Ha detection spectrum, and obtain the hue deviation values ​​from H1 to Ha. S125: The average of the obtained H1 hue deviation values ​​to Ha hue deviation values ​​is calculated to obtain the average hue deviation value corresponding to the pixel points in the sample area. S126: Obtain the hue deviation reference range corresponding to the pixel points in the sample area. If the average hue deviation value corresponding to the pixel points in the sample area is within the hue deviation reference range, then mark the pixel points in the sample area as hue normal pixels. If the average hue deviation value corresponding to the pixel points in the sample area is not within the hue deviation reference range, then mark the pixel points in the sample area as hue abnormal pixels.

3. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 2, characterized in that, Step S12 further includes the following specific steps: S127: Obtain the average hue deviation value corresponding to each region pixel, and divide the obtained region pixels into normal hue pixels and abnormal hue pixels according to the average hue deviation value. S128: Count the number of normal hue pixels in the sample extension sub-region to obtain the number of normal hue pixels. Count the number of abnormal hue pixels in the sample extension sub-region to obtain the number of abnormal hue pixels. Calculate the ratio of the abnormal region area corresponding to the sample extension sub-region by combining the number of normal hue pixels and the number of abnormal hue pixels.

4. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 1, characterized in that, Step S2 includes the following specific steps: S21: Obtain static detection data of the epitaxial structure, obtain the area ratio of the abnormal region corresponding to each sub-region of the epitaxial structure based on the static detection data of the epitaxial structure, and calculate the average value of the obtained multiple abnormal region area ratios to obtain the average value of the abnormal region area ratio. S22: Obtain the acceptable range of abnormal area ratio. If the average value of abnormal area ratio is within the acceptable range of abnormal area ratio, it is determined that the corresponding gallium oxide-based light-emitting diode does not have a static structural defect. If the average value of abnormal area ratio is not within the acceptable range of abnormal area ratio, it is determined that the corresponding gallium oxide-based light-emitting diode has a static structural defect, and static structural detection data is obtained. S23: Perform dynamic epitaxial structure detection on gallium oxide-based light-emitting diodes without static structural defects, and obtain the peak emission brightness deviation based on the detection results; S24: Define the peak emission brightness deviation and static structure detection data as the epitaxial structure dynamic detection data.

5. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 4, characterized in that, Step S23 further includes the following specific steps: S231: In the process of dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, several feature detection powers are set, and a sample feature detection power is selected from the multiple feature detection powers obtained. S232: Perform light emission detection on a gallium oxide-based light-emitting diode at the sample feature detection power to obtain the light emission brightness deviation corresponding to the sample feature detection power; S233: Obtain the luminous brightness deviation corresponding to the detection power of each feature, compare the numerical values ​​of the multiple luminous brightness deviations obtained, and mark the luminous brightness deviation with the largest value as the detection peak luminous brightness deviation.

6. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 5, characterized in that, Step S232 further includes the following specific steps: A gallium oxide-based light-emitting diode is connected to a power-on circuit, and the power of the corresponding power-on circuit is adjusted to the feature detection power. During the dynamic detection of the epitaxial structure of gallium oxide-based light-emitting diodes, the current time point is marked as the end time point of the cycle, the time point when the gallium oxide-based light-emitting diode starts to emit light is marked as the start time point of the cycle, and the time interval between the start time point and the end time point of the cycle is marked as the luminous intensity analysis cycle. The luminous intensity values ​​of gallium oxide-based light-emitting diodes are obtained within the luminous brightness analysis period to obtain the periodic luminous intensity monitoring values. Obtain the reference brightness value of the gallium oxide-based light-emitting diode at the sample feature detection power, calculate the difference between the periodic luminous intensity monitoring value and the reference brightness value, and take the absolute value of the obtained difference to obtain the luminous brightness deviation corresponding to the sample feature detection power.

7. The method for detecting defects in the epitaxial structure of a gallium oxide-based light-emitting diode according to claim 1, characterized in that, Step S3 further includes the following specific steps: S31: Obtain dynamic detection data of the epitaxial structure, and obtain the peak emission brightness deviation and static structure detection data based on the dynamic detection data of the epitaxial structure; S32: If a gallium oxide-based light-emitting diode has static structural defects, a static defect warning will be issued directly for the gallium oxide-based light-emitting diode. S33: If there are no static structural defects in the gallium oxide-based light-emitting diode, then a dynamic defect warning is given to the gallium oxide-based light-emitting diode based on the detected peak luminous brightness deviation; Step S33 further includes the following specific steps: If the detected peak luminous brightness deviation falls within the luminous brightness deviation reference range, then there is no need to issue a dynamic defect warning for the gallium oxide-based light-emitting diode. If the detected peak luminous intensity deviation is not within the luminous intensity deviation reference range, a dynamic defect warning will be issued for the gallium oxide-based light-emitting diode.

Images