Indium phosphide crystal impurity detection method and system based on image processing

Abstract

The present invention relates to the field of crystal impurity detection technology, and in particular to an indium phosphide crystal impurity detection method and system based on image processing. The present invention imports impurity distribution feature data into an impurity distribution difference analysis model to analyze the impurity distribution difference of indium phosphide crystals; imports the impurity distribution difference analysis results and production process control data into a production process difference analysis model to analyze the difference in the production process of indium phosphide crystals; imports the production process difference analysis results into a production process optimization model to optimize the production process of indium phosphide crystals; and adjusts the production parameters of an indium phosphide crystal growth production line according to the production process optimization results of indium phosphide crystals. By optimizing the production parameters of crystal growth according to the characteristics of different impurity categories, the production parameters of crystal growth can be optimized in a targeted manner, which can not only ensure the high efficiency of production, but also maximize the reduction of the generation of specific impurities, thereby improving the purity and performance of indium phosphide crystals.

Description

Technical Field

[0001] The present invention relates to the technical field of crystal impurity detection, and in particular to an indium phosphide crystal impurity detection method and system based on image processing. Background Art

[0002] Indium phosphide crystal is an important semiconductor material, widely used in optoelectronic devices, lasers, fiber optic communications, microwave devices and other fields. The quality of indium phosphide crystal is crucial to the performance of these devices, so the purity and structural integrity of the crystal are very high. The presence of impurities will seriously affect the optical, electrical and mechanical properties of the crystal, causing device performance degradation or failure. Therefore, impurity detection is crucial to improving the quality of indium phosphide crystal.

[0003] Existing impurity detection methods mainly rely on chemical analysis, electrical measurement or physical methods. These methods usually require complex equipment, time-consuming sample preparation and high costs. In addition, many methods are destructive and may damage crystal samples. Moreover, traditional detection methods are inefficient when processing a large number of samples, and it is difficult to meet the needs of efficient and rapid detection in industrial production. With the development of computer vision and image processing technology, the application of image processing in semiconductor detection has gradually received attention. Image processing technology can quickly and non-destructively analyze the impurity distribution on the surface and inside of indium phosphide crystals, obtain the fine structure information of the crystal surface through digital images, and identify and classify impurities based on image features. This method has higher efficiency and automation than traditional detection methods.

[0004] However, when detecting impurities in indium phosphide crystals, the existing technology ignores the impact of different types of impurities on the indium phosphide crystal production process, does not consider the organic association of different types of impurities with various parameters in the production process, and lacks a closed-loop feedback mechanism for adjusting production conditions using impurity classification data. At the same time, the existing technology mostly adopts a "one-size-fits-all" approach to the optimization of the production process of indium phosphide crystal growth production lines, and cannot adopt personalized production process adjustment strategies for different impurity categories, resulting in some impurities that cannot be effectively controlled, affecting the overall quality of the crystal.

[0005] In order to solve these problems, the present application designs an indium phosphide crystal impurity detection method and system based on image processing. Summary of the invention

[0006] In order to overcome the impurities and shortcomings of the prior art, the present invention provides an indium phosphide crystal impurity detection method and system based on image processing, which can optimize the production parameters of crystal growth in a targeted manner according to the characteristics of different impurity categories, thereby ensuring the high efficiency of production and maximizing the reduction of the generation of specific impurities, thereby improving the purity and performance of indium phosphide crystals.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides an indium phosphide crystal impurity detection method based on image processing, comprising the following steps:

[0009] S1. Use a scanning electron microscope on the indium phosphide crystal growth production line to obtain impurity distribution characteristic data of indium phosphide crystals, and at the same time obtain indium phosphide crystal production process control data;

[0010] S2. Importing the impurity distribution characteristic data into the impurity distribution difference analysis model to analyze the impurity distribution difference of the indium phosphide crystal;

[0011] S3, importing the impurity distribution difference analysis results and the production process control data into the production process difference analysis model, and analyzing the difference in the production process of indium phosphide crystals;

[0012] S4, importing the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals;

[0013] S5. According to the results of the production process optimization of indium phosphide crystals, the production parameters of the indium phosphide crystal growth production line are adjusted.

[0014] In an optional implementation manner, step S2 includes the following specific steps:

[0015] S21, obtaining impurity distribution characteristic data of all indium phosphide crystals on the indium phosphide crystal growth production line, classifying all impurity distribution characteristic data of indium phosphide crystals according to the impurity category of the indium phosphide crystals, and obtaining a set of impurity distribution characteristic data of indium phosphide crystals of multiple impurity categories, wherein the impurity distribution characteristic data is an indium phosphide crystal impurity distribution characteristic vector , the impurity distribution feature vector is obtained by processing the impurity distribution image obtained by scanning electron microscope, wherein The impurity texture feature value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity shape characteristic value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity distribution density characteristic value is obtained by processing the impurity distribution image obtained by the scanning electron microscope;

[0016] S22, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category to calculate the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category is:

[0017] ;

[0018] In the formula is the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, k is the number of impurity categories, Ci represents the impurity distribution characteristic data set of indium phosphide crystals of the i-th impurity category, X is the impurity distribution characteristic vector of indium phosphide crystals in the impurity distribution characteristic data set, represents the total number of indium phosphide crystals of the i-th impurity category, Indicates the transposition of the contents in brackets.

[0019] In an optional implementation manner, the step S2 further includes the following specific steps:

[0020] S23, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories to calculate the impurity distribution difference coefficient of indium phosphide crystals of different impurity categories, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories is:

[0021] ;

[0022] In the formula is the impurity distribution difference coefficient of InP crystals with different impurity categories, and N is the total number of InP crystals with all impurity categories;

[0023] S24, obtain the linear discriminant analysis method so that Maximize the discriminant vector matrix W, where is the transpose of the discriminant vector matrix W; the discriminant vectors obtained by the linear discriminant analysis method are arranged in descending order, and the first k-1 discriminant vectors after descending order are composed of the discriminant vector matrix W;

[0024] S25, obtaining the indium phosphide crystal impurity distribution feature vector of the indium phosphide crystal to be classified and importing it into the impurity category classification distance formula to calculate the distance between the impurity distribution feature data of the indium phosphide crystal to be classified and the impurity distribution feature data set of indium phosphide crystals of all impurity categories, wherein the impurity category classification distance formula is:

[0025] ;

[0026] In the formula represents the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of the indium phosphide crystal of the i-th impurity category, represents the impurity distribution feature vector of the indium phosphide crystal to be classified, An indium phosphide crystal impurity distribution feature vector representing the jth indium phosphide crystal in the impurity distribution feature data set of the indium phosphide crystal of the i-th impurity category;

[0027] S26, extracting the impurity category corresponding to the maximum value of the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories as the impurity category of the indium phosphide crystal to be classified.

[0028] In an optional implementation manner, step S3 includes the following specific steps:

[0029] S31, obtaining production process control data of all indium phosphide crystals on the indium phosphide crystal growth production line; the production process control data includes the growth temperature and growth rate of all indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0030] S32, substituting the production process control data into the production process difference coefficient calculation formula to calculate the production process difference coefficient of the indium phosphide crystal to be classified, the production process difference coefficient calculation formula is:

[0031] ;

[0032] Wherein Lc represents the production process difference coefficient of the indium phosphide crystal to be classified, Tz represents the growth temperature difference coefficient of the indium phosphide crystal to be classified, Rz represents the growth rate difference coefficient of the indium phosphide crystal to be classified, a and b are the temperature influence factor and the rate influence factor respectively;

[0033] Among them, the calculation formula of the growth temperature difference coefficient Tz is:

[0034] ;

[0035] Where T is the growth temperature of the indium phosphide crystal to be classified, represents the growth temperature of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0036] The calculation formula of the growth rate difference coefficient Rz is:

[0037] ;

[0038] Where R is the growth rate of the indium phosphide crystal to be classified, represents the growth rate of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0039] S33, preset a production process difference threshold. When the production process difference coefficient of the indium phosphide crystal to be classified is greater than the production process difference threshold, issue an indium phosphide growth production line optimization instruction and execute step S4; when the production process difference coefficient of the indium phosphide crystal to be classified is less than or equal to the production process difference threshold, no instruction is issued.

[0040] In an optional implementation manner, step S4 includes the following specific steps:

[0041] S41, obtaining the calculated production process difference coefficient of the indium phosphide crystals to be classified;

[0042] S42. Construct a production line optimization model through the production process difference coefficient. The function expression of the production line optimization model is:

[0043] ;

[0044] Where F represents the production line optimization model function, min() represents taking the minimum value in the brackets, w1 represents the influence weight of the impurity category corresponding to the indium phosphide crystal to be classified on the production line optimization, exp() represents the exponential function with the natural constant e as the base, Ea is the activation energy of the indium phosphide crystal, Kb is the Boltzmann constant, T is the growth temperature of the indium phosphide crystal to be classified, and w2 represents the influence weight of the growth rate on the production line optimization. Represents the impact weight of production process differences on production line optimization.

[0045] In an optional implementation manner, the step S4 further includes the following specific steps:

[0046] S43. Calculate the partial derivative of the growth temperature in the production line optimization model function, and set the partial derivative to 0 to obtain the optimal growth temperature.

[0047] In an optional implementation manner, step S5 includes the following specific steps:

[0048] S51, obtaining the optimal growth temperature;

[0049] S52, sending the optimal growth temperature as a production parameter to the production line operator, and the production line operator adjusts the indium phosphide crystal growth production line according to the production parameter.

[0050] In a second aspect, an embodiment of the present invention further provides an indium phosphide crystal impurity detection system based on image processing, comprising:

[0051] A data acquisition module is used to obtain the impurity distribution characteristic data of the indium phosphide crystal using a scanning electron microscope on the indium phosphide crystal growth production line, and to obtain the indium phosphide crystal production process control data;

[0052] An impurity distribution difference analysis module is used to import impurity distribution characteristic data into an impurity distribution difference analysis model to analyze the impurity distribution difference of indium phosphide crystals;

[0053] The production process difference analysis module is used to import the impurity distribution difference analysis results and production process control data into the production process difference analysis model to analyze the differences in the production process of indium phosphide crystals;

[0054] The production process optimization analysis module is used to import the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals;

[0055] An indium phosphide crystal growth production line adjustment module is used to adjust the production parameters of the indium phosphide crystal growth production line according to the production process optimization results of the indium phosphide crystal;

[0056] A control module is used to control the operation of the data acquisition module, the impurity distribution difference analysis module, the production process difference analysis module, the production process optimization analysis module and the indium phosphide crystal growth production line adjustment module.

[0057] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes an indium phosphide crystal impurity detection method based on image processing by calling the computer program stored in the memory.

[0058] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing instructions, which, when executed on a computer, enables the computer to execute an indium phosphide crystal impurity detection method based on image processing.

[0059] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0060] The present invention imports impurity distribution characteristic data into an impurity distribution difference analysis model to analyze the impurity distribution difference of indium phosphide crystals; imports the impurity distribution difference analysis results and production process control data into a production process difference analysis model to analyze the difference in the production process of indium phosphide crystals; imports the production process difference analysis results into a production process optimization model to optimize the production process of indium phosphide crystals; and adjusts the production parameters of the indium phosphide crystal growth production line according to the production process optimization results of the indium phosphide crystals. By optimizing the production parameters of the crystal growth line in a targeted manner according to the characteristics of different impurity categories, the production efficiency of the production can be guaranteed, and the generation of specific impurities can be minimized, thereby improving the purity and performance of the indium phosphide crystals. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0062] Figure 1 It is a schematic diagram of the overall process of the indium phosphide crystal impurity detection method based on image processing of the present invention;

[0063] Figure 2 A production flow chart of indium phosphide crystals in the indium phosphide crystal impurity detection method based on image processing of the present invention;

[0064] Figure 3 It is a schematic structural diagram of the indium phosphide crystal impurity detection system based on image processing of the present invention. DETAILED DESCRIPTION

[0065] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0066] Example 1

[0067] like Figure 2 As shown, this embodiment provides a production process of indium phosphide crystals, which specifically includes the following stages: a material preparation stage, which is used to prepare high-purity raw materials for the growth of indium phosphide crystals; an indium phosphide seed preparation stage, which is used to prepare high-quality indium phosphide seed crystals and guide subsequent crystal growth; a crystal growth stage, which is used to combine indium and phosphorus under controlled conditions to generate indium phosphide, and guide it to grow along a specific crystal direction through a seed crystal to form a large indium phosphide single crystal; a crystal processing stage, which is used to cut, grind and polish the grown indium phosphide crystals to obtain wafers of required shape and size; a quality inspection stage, which is used to perform quality inspection on crystals and wafers to ensure that the wafers meet the requirements of device production; a doping and epitaxial growth stage, which is used to give the crystals specific electrical properties through doping and epitaxial processes, and grow the required thin layer on the wafer surface; a device processing stage, which is used to process the doped and epitaxial indium phosphide wafers into specific semiconductor devices, such as photodiodes, lasers, etc.

[0068] It should be noted that the image processing-based indium phosphide crystal impurity detection method provided in this embodiment is used to detect the impurity types in the indium phosphide crystal growth stage, and to optimize the production parameters of the indium phosphide crystal growth stage in a targeted manner according to the characteristics of different impurity types.

[0069] like Figure 1As shown, this embodiment provides an indium phosphide crystal impurity detection method based on image processing, which specifically includes the following steps:

[0070] S1. Use a scanning electron microscope on the indium phosphide crystal growth production line to obtain impurity distribution characteristic data of indium phosphide crystals, and at the same time obtain indium phosphide crystal production process control data;

[0071] S2. Importing the impurity distribution characteristic data into the impurity distribution difference analysis model to analyze the impurity distribution difference of the indium phosphide crystal;

[0072] In this embodiment, step S2 includes the following specific steps:

[0073] S21, obtaining impurity distribution characteristic data of all indium phosphide crystals on the indium phosphide crystal growth production line, classifying all impurity distribution characteristic data of indium phosphide crystals according to the impurity category of the indium phosphide crystals, and obtaining a set of impurity distribution characteristic data of indium phosphide crystals of multiple impurity categories, wherein the impurity distribution characteristic data is an indium phosphide crystal impurity distribution characteristic vector , the impurity distribution feature vector is obtained by processing the impurity distribution image obtained by scanning electron microscope, wherein The impurity texture feature value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity shape characteristic value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity distribution density characteristic value is obtained by processing the impurity distribution image obtained by the scanning electron microscope;

[0074] It should be noted that in order to extract impurity texture feature values, impurity shape feature values ​​and impurity distribution density feature values ​​from the impurity distribution image obtained by a scanning electron microscope (SEM), the following image processing technology methods can be used:

[0075] In this embodiment, the impurity texture feature value can be obtained by analyzing the direction and distance relationship between pixel pairs in the impurity distribution image through the gray level co-occurrence matrix, so as to calculate the texture feature of the impurity distribution image; the texture feature of the impurity distribution image can also be analyzed at different scales and directions through wavelet transform. The impurity shape feature value can be obtained by detecting the edge in the image through edge detection operators such as Canny and Sobel, thereby extracting the contour shape features of the impurities; for impurities with regular shapes, the shape parameters of the impurities in the impurity distribution image can be effectively identified through Hough transform. In this embodiment, the impurity distribution image can also be divided into different regions according to the similarity of pixels in the impurity distribution image through an image segmentation algorithm, so as to identify the shape and size of the impurities. In this embodiment, one or more thresholds can be set to obtain the impurity distribution density feature value, and the pixels in the image can be divided into two categories: impurities and non-impurities, so as to calculate the number and density of impurities. For a large number of discrete impurity points, a point pattern analysis method can be used to study the spatial distribution pattern of impurities, such as the nearest neighbor distance, Ripley's K function, etc. It should be noted that the above-mentioned image processing technology methods are all prior art and will not be repeated here.

[0076] S22, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category to calculate the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category is:

[0077] ;

[0078] In the formula is the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, k is the number of impurity categories, Ci represents the impurity distribution characteristic data set of indium phosphide crystals of the i-th impurity category, X is the impurity distribution characteristic vector of indium phosphide crystals in the impurity distribution characteristic data set, represents the total number of indium phosphide crystals of the i-th impurity category, Indicates the transposition of the contents in brackets.

[0079] In this embodiment, step S2 also includes the following specific steps:

[0080] S23, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories to calculate the impurity distribution difference coefficient of indium phosphide crystals of different impurity categories, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories is:

[0081] ;

[0082] In the formula is the impurity distribution difference coefficient of InP crystals with different impurity categories, and N is the total number of InP crystals with all impurity categories;

[0083] S24, obtain the linear discriminant analysis method so that Maximize the discriminant vector matrix W, where is the transpose of the discriminant vector matrix W; the discriminant vectors obtained by the linear discriminant analysis method are arranged in descending order, and the first k-1 discriminant vectors after descending order are composed of the discriminant vector matrix W;

[0084] It should be noted that the linear discriminant analysis method optimizes data projection by maximizing the divergence between impurity categories and minimizing the divergence within the category, so that indium phosphide crystals of different impurity categories are separated as much as possible in the low-dimensional space. In the present embodiment, when the impurities of indium phosphide crystals are classified, the linear discriminant analysis method can effectively separate different types of impurities, such as metal impurities, non-metallic impurities, etc. At the same time, the linear discriminant analysis method can also project the high-dimensional feature vector of the scanning electron microscope image into the low-dimensional space, while retaining important impurity category distinction information. The linear discriminant analysis method is of great significance for processing high-dimensional data, which can reduce computational complexity and storage requirements. In the present embodiment, the impurity characteristics of indium phosphide crystals are reduced from tens of dimensions or hundreds of dimensions to a few dimensions, which is convenient for subsequent analysis and processing. The linear discriminant analysis method can use category labels to guide the construction of feature space, and can optimize the representation of data according to category information, thereby improving the classification effect. In the present embodiment, the feature vectors of indium phosphide crystal impurities processed by the linear discriminant analysis method will have a more obvious distinction between categories, thereby improving the recognition accuracy of different impurity categories.

[0085] S25, obtaining the indium phosphide crystal impurity distribution feature vector of the indium phosphide crystal to be classified and importing it into the impurity category classification distance formula to calculate the distance between the impurity distribution feature data of the indium phosphide crystal to be classified and the impurity distribution feature data set of indium phosphide crystals of all impurity categories, wherein the impurity category classification distance formula is:

[0086] ;

[0087] In the formula represents the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of the indium phosphide crystal of the i-th impurity category, represents the impurity distribution feature vector of the indium phosphide crystal to be classified, An indium phosphide crystal impurity distribution feature vector representing the jth indium phosphide crystal in the impurity distribution feature data set of the indium phosphide crystal of the i-th impurity category;

[0088] S26, extracting the impurity category corresponding to the maximum value of the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories as the impurity category of the indium phosphide crystal to be classified.

[0089] S3, importing the impurity distribution difference analysis results and the production process control data into the production process difference analysis model, and analyzing the difference in the production process of indium phosphide crystals;

[0090] In this embodiment, step S3 includes the following specific steps:

[0091] S31, obtaining production process control data of all indium phosphide crystals on the indium phosphide crystal growth production line; the production process control data includes the growth temperature and growth rate of all indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0092] S32, substituting the production process control data into the production process difference coefficient calculation formula to calculate the production process difference coefficient of the indium phosphide crystal to be classified, the production process difference coefficient calculation formula is:

[0093] ;

[0094] Wherein Lc represents the production process difference coefficient of the indium phosphide crystal to be classified, Tz represents the growth temperature difference coefficient of the indium phosphide crystal to be classified, Rz represents the growth rate difference coefficient of the indium phosphide crystal to be classified, a and b are the temperature influence factor and the rate influence factor, respectively. It should be noted that the above influence factors can be set by those skilled in the art according to the actual situation of the indium phosphide crystal growth production line and the specific needs of the indium phosphide crystal;

[0095] Among them, the calculation formula of the growth temperature difference coefficient Tz is:

[0096] ;

[0097] Where T is the growth temperature of the indium phosphide crystal to be classified, represents the growth temperature of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0098] The calculation formula of the growth rate difference coefficient Rz is:

[0099] ;

[0100] Where R is the growth rate of the indium phosphide crystal to be classified, represents the growth rate of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0101] S33, preset a production process difference threshold. When the production process difference coefficient of the indium phosphide crystal to be classified is greater than the production process difference threshold, issue an indium phosphide growth production line optimization instruction and execute step S4. When the production process difference coefficient of the indium phosphide crystal to be classified is less than or equal to the production process difference threshold, no instruction is issued. It should be noted that the production process difference threshold is set by technicians in this field according to actual needs and specific circumstances.

[0102] S4, importing the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals;

[0103] In this embodiment, step S4 includes the following specific steps:

[0104] S41, obtaining the calculated production process difference coefficient of the indium phosphide crystals to be classified;

[0105] S42. Construct a production line optimization model through the production process difference coefficient. The function expression of the production line optimization model is:

[0106] ;

[0107] Wherein F represents the production line optimization model function, min() represents taking the minimum value in the brackets, w1 represents the influence weight of the impurity category corresponding to the indium phosphide crystal to be classified on the production line optimization, exp() represents the exponential function with the natural constant e as the base, Ea is the activation energy of the indium phosphide crystal. It should be noted that the activation energy of the indium phosphide crystal refers to the energy required for the diffusion of impurities in the indium phosphide crystal, which is usually between 0.6 eV and 1.2 eV. Those skilled in the art can set the activation energy of the indium phosphide crystal according to actual conditions; Kb is the Boltzmann constant, T is the growth temperature of the indium phosphide crystal to be classified, w2 represents the influence weight of the growth rate on the production line optimization, It represents the influence weight of production process difference on production line optimization. It should be noted that the above influence weight can be set by technical personnel in this field according to the actual situation of the indium phosphide crystal growth production line and the specific demand for indium phosphide crystals.

[0108] In this embodiment, step S4 also includes the following specific steps:

[0109] S43. Calculate the partial derivative of the growth temperature in the production line optimization model function, and set the partial derivative to 0 to obtain the optimal growth temperature.

[0110] S5. According to the results of the production process optimization of indium phosphide crystals, the production parameters of the indium phosphide crystal growth production line are adjusted.

[0111] In this embodiment, step S5 includes the following specific steps:

[0112] S51, obtaining the optimal growth temperature;

[0113] S52, sending the optimal growth temperature as a production parameter to the production line operator, and the production line operator adjusts the indium phosphide crystal growth production line according to the production parameter.

[0114] Example 2

[0115] like Figure 3 As shown, this embodiment provides an indium phosphide crystal impurity detection system based on image processing, comprising:

[0116] A data acquisition module is used to obtain the impurity distribution characteristic data of the indium phosphide crystal using a scanning electron microscope on the indium phosphide crystal growth production line, and to obtain the indium phosphide crystal production process control data;

[0117] An impurity distribution difference analysis module is used to import impurity distribution characteristic data into an impurity distribution difference analysis model to analyze the impurity distribution difference of indium phosphide crystals;

[0118] The production process difference analysis module is used to import the impurity distribution difference analysis results and production process control data into the production process difference analysis model to analyze the differences in the production process of indium phosphide crystals;

[0119] The production process optimization analysis module is used to import the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals;

[0120] An indium phosphide crystal growth production line adjustment module is used to adjust the production parameters of the indium phosphide crystal growth production line according to the production process optimization results of the indium phosphide crystal;

[0121] The control module is used to control the operation of the data acquisition module, the impurity distribution difference analysis module, the production process difference analysis module, the production process optimization analysis module and the indium phosphide crystal growth production line adjustment module.

[0122] In this embodiment, the impurity distribution difference analysis module is used to import the impurity distribution characteristic data into the impurity distribution difference analysis model to analyze the impurity distribution difference of the indium phosphide crystal; the analysis of the impurity distribution difference of the indium phosphide crystal includes the following specific steps:

[0123] Obtain impurity distribution characteristic data of all indium phosphide crystals on the indium phosphide crystal growth production line, classify all impurity distribution characteristic data of indium phosphide crystals according to the impurity category of the indium phosphide crystals, and obtain a set of impurity distribution characteristic data of indium phosphide crystals of multiple impurity categories, wherein the impurity distribution characteristic data is an indium phosphide crystal impurity distribution characteristic vector , the impurity distribution feature vector is obtained by processing the impurity distribution image obtained by scanning electron microscope, wherein The impurity texture feature value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity shape characteristic value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity distribution density characteristic value is obtained by processing the impurity distribution image obtained by the scanning electron microscope;

[0124] Substitute the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category to calculate the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category. The impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category is:

[0125] ;

[0126] In the formula is the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, k is the number of impurity categories, Ci represents the impurity distribution characteristic data set of indium phosphide crystals of the i-th impurity category, X is the impurity distribution characteristic vector of indium phosphide crystals in the impurity distribution characteristic data set, represents the total number of indium phosphide crystals of the i-th impurity category, Indicates the transposition of the contents in brackets.

[0127] In this embodiment, analyzing the difference in impurity distribution of indium phosphide crystals further includes the following specific steps:

[0128] Substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories to calculate the impurity distribution difference coefficient of indium phosphide crystals of different impurity categories, the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories is:

[0129] ;

[0130] In the formula is the impurity distribution difference coefficient of InP crystals with different impurity categories, and N is the total number of InP crystals with all impurity categories;

[0131] The linear discriminant analysis method is used to obtain Maximize the discriminant vector matrix W, where is the transpose of the discriminant vector matrix W; the discriminant vectors obtained by the linear discriminant analysis method are arranged in descending order, and the first k-1 discriminant vectors after descending order are composed of the discriminant vector matrix W;

[0132] The impurity distribution feature vector of the indium phosphide crystal to be classified is obtained and imported into the impurity category classification distance formula to calculate the distance between the impurity distribution feature data of the indium phosphide crystal to be classified and the impurity distribution feature data set of indium phosphide crystals of all impurity categories. The impurity category classification distance formula is:

[0133] ;

[0134] In the formula represents the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of the indium phosphide crystal of the i-th impurity category, represents the impurity distribution feature vector of the indium phosphide crystal to be classified, An indium phosphide crystal impurity distribution feature vector representing the jth indium phosphide crystal in the impurity distribution feature data set of the indium phosphide crystal of the i-th impurity category;

[0135] The impurity category corresponding to the maximum value of the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories is extracted as the impurity category of the indium phosphide crystal to be classified.

[0136] In this embodiment, the production process difference analysis module is used to import the impurity distribution difference analysis results and the production process control data into the production process difference analysis model to analyze the difference in the production process of indium phosphide crystals; the analysis of the difference in the production process of indium phosphide crystals includes the following specific steps:

[0137] Obtaining production process control data of all indium phosphide crystals on an indium phosphide crystal growth production line; the production process control data includes growth temperature and growth rate of all indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0138] Substitute the production process control data into the production process difference coefficient calculation formula to calculate the production process difference coefficient of the indium phosphide crystal to be classified. The production process difference coefficient calculation formula is:

[0139] ;

[0140] Wherein Lc represents the production process difference coefficient of the indium phosphide crystal to be classified, Tz represents the growth temperature difference coefficient of the indium phosphide crystal to be classified, Rz represents the growth rate difference coefficient of the indium phosphide crystal to be classified, a and b are the temperature influence factor and the rate influence factor, respectively. It should be noted that the above influence factors can be set by those skilled in the art according to the actual situation of the indium phosphide crystal growth production line and the specific needs of the indium phosphide crystal;

[0141] Among them, the calculation formula of the growth temperature difference coefficient Tz is:

[0142] ;

[0143] Where T is the growth temperature of the indium phosphide crystal to be classified, represents the growth temperature of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0144] The calculation formula of the growth rate difference coefficient Rz is:

[0145] ;

[0146] Where R is the growth rate of the indium phosphide crystal to be classified, represents the growth rate of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified;

[0147] A production process difference threshold is preset. When the production process difference coefficient of the indium phosphide crystal to be classified is greater than the production process difference threshold, an indium phosphide growth production line optimization instruction is issued and step S4 is executed; when the production process difference coefficient of the indium phosphide crystal to be classified is less than or equal to the production process difference threshold, no instruction is issued. It should be noted that the production process difference threshold is set by technical personnel in this field according to actual needs and specific circumstances.

[0148] In this embodiment, the production process optimization analysis module is used to import the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals;

[0149] In this embodiment, optimizing the production process of indium phosphide crystals includes the following specific steps:

[0150] Obtain the calculated production process difference coefficient of the indium phosphide crystal to be classified;

[0151] The production line optimization model is constructed by using the production process difference coefficient. The function expression of the production line optimization model is:

[0152] ;

[0153] Where F represents the production line optimization model function, min() represents taking the minimum value in the brackets, w1 represents the influence weight of the impurity category corresponding to the indium phosphide crystal to be classified on the production line optimization, exp() represents the exponential function with the natural constant e as the base, Ea is the activation energy of the indium phosphide crystal, Kb is the Boltzmann constant, T is the growth temperature of the indium phosphide crystal to be classified, and w2 represents the influence weight of the growth rate on the production line optimization. It represents the influence weight of production process difference on production line optimization. It should be noted that the above influence weight can be set by technical personnel in this field according to the actual situation of the indium phosphide crystal growth production line and the specific demand for indium phosphide crystals.

[0154] In this embodiment, optimizing the production process of indium phosphide crystals further includes the following specific steps:

[0155] The partial derivative of the growth temperature in the production line optimization model function is calculated and set to 0 to obtain the optimal growth temperature.

[0156] In this embodiment, the indium phosphide crystal growth production line adjustment module is used to adjust the production parameters of the indium phosphide crystal growth production line according to the production process optimization result of the indium phosphide crystal; the production parameters of the indium phosphide crystal growth production line are adjusted, including the following specific steps:

[0157] Obtain optimal growth temperature;

[0158] The optimal growth temperature is sent as a production parameter to the production line operator, who adjusts the indium phosphide crystal growth line according to the production parameter.

[0159] The above-mentioned parameters in the indium phosphide crystal impurity detection system based on image processing of the present invention and the steps for each unit module to achieve corresponding functions can refer to the parameters and steps in the embodiment of the indium phosphide crystal impurity detection method based on image processing above, and will not be repeated here.

[0160] Example 3

[0161] An electronic device according to an embodiment of the present invention includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes an indium phosphide crystal impurity detection method based on image processing by calling the computer program stored in the memory. It should be noted that all computer programs of the indium phosphide crystal impurity detection method based on image processing are implemented in C language, wherein the data acquisition module, the impurity distribution difference analysis module, the production process difference analysis module, the production process optimization analysis module, the indium phosphide crystal growth production line adjustment module and the control module are all controlled by a remote server.

[0162] Example 4

[0163] This embodiment provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0164] When the computer program is executed on a computer device, the computer device is enabled to execute the above-mentioned indium phosphide crystal impurity detection method based on image processing.

[0165] Each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the IoT device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0166] The system and medium provided in the embodiments of the present invention correspond one-to-one to the method, and therefore, the system and medium also have similar beneficial technical effects to the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the system and medium will not be repeated here.

[0167] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0168] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0169] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0170] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0171] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0172] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0173] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0174] The above are only embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An indium phosphide crystal impurity detection method based on image processing, characterized in that: The steps include: S1. Use a scanning electron microscope on the indium phosphide crystal growth production line to obtain impurity distribution characteristic data of indium phosphide crystals, and at the same time obtain indium phosphide crystal production process control data; S2, importing the impurity distribution characteristic data into the impurity distribution difference analysis model to analyze the impurity distribution difference of the indium phosphide crystal; S3, importing the impurity distribution difference analysis results and the production process control data into the production process difference analysis model, and analyzing the difference in the production process of indium phosphide crystals; S4, importing the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals; S5. According to the optimization results of the production process of indium phosphide crystals, the production parameters of the indium phosphide crystal growth production line are adjusted; The step S2 comprises the following specific steps: S21, obtaining impurity distribution characteristic data of all indium phosphide crystals on the indium phosphide crystal growth production line, classifying all impurity distribution characteristic data of indium phosphide crystals according to the impurity category of the indium phosphide crystals, and obtaining a set of impurity distribution characteristic data of indium phosphide crystals of multiple impurity categories, wherein the impurity distribution characteristic data is an indium phosphide crystal impurity distribution characteristic vector , the impurity distribution feature vector is obtained by processing the impurity distribution image obtained by scanning electron microscope, wherein The impurity texture feature value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity shape characteristic value is obtained by processing the impurity distribution image obtained by scanning electron microscope. The impurity distribution density characteristic value is obtained by processing the impurity distribution image obtained by the scanning electron microscope; S22, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category to calculate the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of the same impurity category is: ； In the formula is the impurity distribution difference coefficient of indium phosphide crystals of the same impurity category, k is the number of impurity categories, Ci represents the impurity distribution characteristic data set of indium phosphide crystals of the i-th impurity category, X is the impurity distribution characteristic vector of indium phosphide crystals in the impurity distribution characteristic data set, represents the total number of indium phosphide crystals of the i-th impurity category, Indicates the transposition of the contents in brackets; The step S2 also includes the following specific steps: S23, substituting the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories into the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories to calculate the impurity distribution difference coefficient of indium phosphide crystals of different impurity categories, wherein the impurity distribution difference coefficient calculation formula of indium phosphide crystals of different impurity categories is: ； In the formula is the impurity distribution difference coefficient of indium phosphide crystals with different impurity categories, and N is the total number of indium phosphide crystals with all impurity categories; S24, obtain the linear discriminant analysis method so that Maximize the discriminant vector matrix W, where is the transpose of the discriminant vector matrix W; The discriminant vectors obtained by the linear discriminant analysis method are arranged in descending order, and the first k-1 discriminant vectors after descending order are formed into a discriminant vector matrix W; S25, obtaining the indium phosphide crystal impurity distribution feature vector of the indium phosphide crystal to be classified and importing it into the impurity category classification distance formula to calculate the distance between the impurity distribution feature data of the indium phosphide crystal to be classified and the impurity distribution feature data set of indium phosphide crystals of all impurity categories, wherein the impurity category classification distance formula is: ； In the formula represents the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of the indium phosphide crystal of the i-th impurity category, represents the impurity distribution feature vector of the indium phosphide crystal to be classified, An indium phosphide crystal impurity distribution feature vector representing the jth indium phosphide crystal in the impurity distribution feature data set of the indium phosphide crystal of the i-th impurity category; S26, extracting the impurity category corresponding to the maximum value of the distance between the impurity distribution characteristic data of the indium phosphide crystal to be classified and the impurity distribution characteristic data set of indium phosphide crystals of all impurity categories as the impurity category of the indium phosphide crystal to be classified; The step S3 comprises the following specific steps: S31, obtaining production process control data of all indium phosphide crystals on the indium phosphide crystal growth production line; the production process control data includes the growth temperature and growth rate of all indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified; S32, substituting the production process control data into the production process difference coefficient calculation formula to calculate the production process difference coefficient of the indium phosphide crystal to be classified, the production process difference coefficient calculation formula is: ； Wherein Lc represents the production process difference coefficient of the indium phosphide crystal to be classified, Tz represents the growth temperature difference coefficient of the indium phosphide crystal to be classified, Rz represents the growth rate difference coefficient of the indium phosphide crystal to be classified, a and b are the temperature influence factor and the rate influence factor respectively; Among them, the calculation formula of the growth temperature difference coefficient Tz is: ； Where T is the growth temperature of the indium phosphide crystal to be classified, represents the growth temperature of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified; The calculation formula of the growth rate difference coefficient Rz is: ； Where R is the growth rate of the indium phosphide crystal to be classified, represents the growth rate of the qth indium phosphide crystal in the impurity category corresponding to the indium phosphide crystal to be classified, q is any one of 1 to Q, and Q is the number of indium phosphide crystals in the impurity category corresponding to the indium phosphide crystal to be classified; S33, preset a production process difference threshold, when the production process difference coefficient of the indium phosphide crystal to be classified is greater than the production process difference threshold, issue an indium phosphide growth production line optimization instruction, and execute step S4; when the production process difference coefficient of the indium phosphide crystal to be classified is less than or equal to the production process difference threshold, no instruction is issued; The step S4 comprises the following specific steps: S41, obtaining the calculated production process difference coefficient of the indium phosphide crystals to be classified; S42. Construct a production line optimization model through the production process difference coefficient. The function expression of the production line optimization model is: ； Where F represents the production line optimization model function, min() represents taking the minimum value in the brackets, w1 represents the influence weight of the impurity category corresponding to the indium phosphide crystal to be classified on the production line optimization, exp() represents the exponential function with the natural constant e as the base, Ea is the activation energy of the indium phosphide crystal, Kb is the Boltzmann constant, T is the growth temperature of the indium phosphide crystal to be classified, and w2 represents the influence weight of the growth rate on the production line optimization. Represents the impact weight of production process differences on production line optimization.

2. The method for detecting impurities in indium phosphide crystals based on image processing according to claim 1, characterized in that: The step S4 also includes the following specific steps: S43. Calculate the partial derivative of the growth temperature in the production line optimization model function, and set the partial derivative to 0 to obtain the optimal growth temperature.

3. The method for detecting impurities in indium phosphide crystals based on image processing according to claim 2, characterized in that: The step S5 comprises the following specific steps: S51, obtaining the optimal growth temperature; S52, sending the optimal growth temperature as a production parameter to the production line operator, and the production line operator adjusts the indium phosphide crystal growth production line according to the production parameter.

4. An indium phosphide crystal impurity detection system based on image processing, which is implemented based on the indium phosphide crystal impurity detection method based on image processing according to any one of claims 1 to 3, characterized in that: The system comprises: A data acquisition module is used to obtain the impurity distribution characteristic data of the indium phosphide crystal using a scanning electron microscope on the indium phosphide crystal growth production line, and to obtain the indium phosphide crystal production process control data; An impurity distribution difference analysis module is used to import impurity distribution characteristic data into an impurity distribution difference analysis model to analyze the impurity distribution difference of indium phosphide crystals; The production process difference analysis module is used to import the impurity distribution difference analysis results and production process control data into the production process difference analysis model to analyze the differences in the production process of indium phosphide crystals; The production process optimization analysis module is used to import the production process difference analysis results into the production process optimization model to optimize the production process of indium phosphide crystals; An indium phosphide crystal growth production line adjustment module is used to adjust the production parameters of the indium phosphide crystal growth production line according to the production process optimization results of the indium phosphide crystal; A control module is used to control the operation of the data acquisition module, the impurity distribution difference analysis module, the production process difference analysis module, the production process optimization analysis module and the indium phosphide crystal growth production line adjustment module.

5. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the indium phosphide crystal impurity detection method based on image processing as described in any one of claims 1 to 3 by calling the computer program stored in the memory.

6. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the indium phosphide crystal impurity detection method based on image processing as described in any one of claims 1 to 3.

Images