A processing method, system and storage medium for Czochralski crystal growth images

The direct-pull crystal growth images are processed through the Zero-DCE algorithm, dark channel prior algorithm and adaptive gamma correction algorithm, which solves the image noise and halo problems, improves the image contrast and brightness, and improves the monitoring effect of the crystal growth process.

CN118037723BActive Publication Date: 2024-11-12QUANZHOU INST OF EQUIP MFG +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410434323.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-11
Publication Date
2024-11-12
Estimated Expiration
2044-04-11

AI Technical Summary

Technical Problem

During the growth of straight-pulfrost crystals, due to high temperature and bright environments, images collected by industrial cameras are prone to noise and halos, resulting in low image contrast and it is difficult to effectively monitor crystal growth.

Method used

The Zero-DCE algorithm is used to segment and splice the image, combined with dark channel prior algorithm and adaptive gamma correction algorithm, to remove halo and noise, and improve image brightness and contrast.

Benefits of technology

The image quality is significantly improved, so that the processed images can better assist in the manual observation of the crystal growth process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118037723B_ABST
    Figure CN118037723B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of image processing, and specifically discloses a method, system and storage medium for processing images of Czochralski crystal growth, comprising the following steps: S1: acquiring an original image #imgabs0#; S2: segmenting the original image #imgabs1# three times to obtain a third layer of sub-images #imgabs2#, using a Zero‑DCE algorithm to process the third layer of sub-images #imgabs3# and splicing four adjacent sub-images together, performing a recursive operation until a new image #imgabs4# is obtained; S3: performing a window moving operation three times on the new image #imgabs5# S4: calculate the transmittance in the single crystal furnace #imgabs9#, and use the dark channel prior algorithm to calculate the observed image #imgabs10#; S5: process the observed image #imgabs11#; S6: output the image processed in step S5; improve the image quality so as to better assist manual observation of the crystal growth process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a method, system and storage medium for processing images of Czochralski crystal growth. Background Art

[0002] The Czochralski method is a crystal growth method that is very suitable for industrial production. It is widely used in single crystals used in modern technologies such as electronics and optics. However, the entire process of Czochralski crystal growth is always in a high temperature and high brightness environment. In the past, workers had to rely on their naked eyes to continuously observe the growth process (which may last for several days), which is very harmful to the human eye. With the development of machine vision technology, people began to try to use industrial cameras instead of human eyes to monitor the crystal growth process. However, due to the unique growth environment, the imaging of crystals by industrial cameras is often accompanied by many noise points and halos. In addition, due to the high light transmittance and reflective properties of the crystal itself, the image contrast received by the industrial camera is generally low, and the effect of directly using it for crystal growth observation is not good. Summary of the invention

[0003] The object of the present invention is to provide a method, system and storage medium for processing Czochralski crystal growth images which can improve the image quality.

[0004] In order to achieve the above object, the present invention adopts such technical solution:

[0005] A method for processing a Czochralski crystal growth image comprises the following processing steps performed in sequence:

[0006] S1: Collecting raw images during crystal growth ;

[0007] S2: The original image The first layer sub-image is split into 4 * 4 , the first layer sub-image The second layer sub-image is divided into 8 * 8 , the second layer sub-image Split into 16 * 16 third layer sub-images , the Zero-DCE algorithm is used to extract the third layer sub-image Process and stitch the four adjacent sub-images together to get the second sub-image of 8 * 8 , the Zero-DCE algorithm is used to extract the second sub-image Process and stitch the four adjacent sub-images together to get the first sub-image of 4 * 4 , use Zero-DCE algorithm to calculate the first sub-image Process and stitch four adjacent sub-images together to get a new image ;

[0008] S3: Segment the new image according to a preset segmentation form Perform the window moving operation and splitting operation three times in sequence to obtain a new third-layer sub-image , use Zero-DCE algorithm to extract the new sub-image After processing, a splicing operation is performed according to the preset segmentation form to obtain a new second layer sub-image , and then recursively operate until a new first image is obtained ;

[0009] S4: Solving the first image Dark channel , from the dark channel Select a preset ratio of points from the image and correspond to the first image To find the mean of these points, as the first image The global estimate of , calculate the transmittance in the single crystal furnace according to the temperature and pressure in the single crystal furnace , the dark channel prior algorithm is used to calculate the observed image ;

[0010] S5: Adaptive gamma method for observing images Perform brightness enhancement processing to transform the observed image Convert to Space, extract observation image Brightness information ,Keep and The channel remains unchanged, and the brightness information To enhance, the formula for brightness information enhancement is as follows:

[0011] ;

[0012] In the formula, is the image brightness information after gamma enhancement, is any pixel in the image, is the gamma correction factor, The expression is as follows:

[0013] ;

[0014] In the formula, is the maximum value of the brightness channel; is the minimum value of the brightness channel;

[0015] S6: Output the image processed in step S5.

[0016] Preferably, in step S1, an industrial camera with an ND filter on the lens is used to obtain the original image. Collection of.

[0017] A system for processing images of Czochralski crystal growth comprises a memory and a processor, wherein the memory stores executable commands of the processor, and the processor is configured to implement any of the above-mentioned methods by executing the executable commands.

[0018] A storage medium stores a computer program, which implements any of the methods described above when executed.

[0019] By adopting the above-mentioned design scheme, the beneficial effects of the present invention are as follows: the present invention adopts the Zero-DCE algorithm, the dark channel prior algorithm based on guided filtering and the adaptive gamma correction algorithm to enhance the image data generated in the Czochralski crystal growth process, improve the brightness, remove the halo and noise, and improve the image contrast, so as to improve the image quality, so that the processed image can better assist manual observation of the crystal growth process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is the network architecture obtained by training the Zero-DCE algorithm;

[0021] Figure 2 It is the Zero-DCE algorithm flow for multi-scale processing combined with window movement;

[0022] Figure 3 It is a schematic diagram of some collected original images and processed images;

[0023] Figure 2 In the figure, Figure a is a flow chart of the segmentation of the original image L in step S2, and Figure b is a flow chart of the window movement and segmentation operation in step S3. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only partial embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] A method for processing a Czochralski crystal growth image comprises the following processing steps performed in sequence:

[0026] S1: Collecting raw images during crystal growth ( Figure 3 The following are some examples of the original images collected);

[0027] In this embodiment, in step S1, an industrial camera with an ND filter on the lens is used to obtain the original image. It can also collect video data and then convert the video data frame by frame into image processing.

[0028] like Figure 1 As shown, S2: Figure 2 As shown in Figure a, the original image The first layer sub-image is split into 4 * 4 , the first layer sub-image The second layer sub-image is divided into 8 * 8 , the second layer sub-image Split into 16 * 16 third layer sub-images , the Zero-DCE algorithm is used to extract the third layer sub-image Process and stitch the four adjacent sub-images together to get the second sub-image of 8 * 8 , the Zero-DCE algorithm is used to extract the second sub-image Process and stitch the four adjacent sub-images together to get the first sub-image of 4 * 4 , use Zero-DCE algorithm to calculate the first sub-image Process and stitch four adjacent sub-images together to get a new image ;

[0029] like Figure 2 As shown, S3: the new image is segmented according to the preset segmentation form Perform the window moving operation and splitting operation three times in sequence to obtain a new third-layer sub-image , use Zero-DCE algorithm to extract the new sub-image After processing, the splicing operation is performed according to the preset segmentation form to obtain a new second-layer sub-image , recursively operate until the first image is obtained ;

[0030] In this embodiment, the Figure 2 The preset segmentation form shown in Figure b is used for the new image Perform window moving and splitting operations, and you can also make other settings according to actual needs. The subsequent splicing operation also uses the preset splitting form shown in Figure b to perform splicing and restoration.

[0031] S4: Solving the first image Dark channel , from the dark channel A preset ratio of points is selected from the first image. In this embodiment, 0.1% is used as the preset ratio. The preset ratio can also be set according to actual needs. Find the mean of these points as the first image The global estimate of , calculate the transmittance in the single crystal furnace according to the temperature and pressure in the single crystal furnace , the dark channel prior algorithm is used to calculate the observed image ;

[0032] In this embodiment, the brightness of the first image is calculated using the following formula:

[0033] ;

[0034] in, It refers to the observed brightness, that is, the brightness obtained from the captured picture; The image after dehalation and restoration; is the transmittance; is the global estimate;

[0035] The observed image is calculated using the following formula: Dark channel:

[0036] ;

[0037] in, express Any color channel of Indicated in pixels The window is , and c represents the image channel.

[0038] S5: Adaptive gamma for observed images Perform brightness enhancement processing to transform the observed image Convert to Space, extract observation image Brightness information ,Keep and The channel remains unchanged, and the brightness information To enhance, the formula for brightness information enhancement is as follows:

[0039] ;

[0040] In the formula, is the image brightness information after gamma enhancement, is any pixel in the image, is the gamma correction factor, The expression is as follows:

[0041] ;

[0042] In the formula, is the maximum value of the brightness channel; is the minimum value of the brightness channel;

[0043] S6: Output the image processed in step S5.

[0044] This embodiment also provides a system capable of implementing the above method.

[0045] A system for processing images of Czochralski crystal growth comprises a memory and a processor, wherein the memory stores executable commands of the processor, and the processor is configured to implement any of the above-mentioned methods by executing the executable commands.

[0046] This embodiment also provides a storage medium storing a computer program capable of implementing the above method.

[0047] A storage medium stores a computer program, which implements any of the methods described above when executed.

[0048] In summary, the present invention adopts the Zero-DCE algorithm, the dark channel prior algorithm based on guided filtering and the adaptive gamma correction algorithm to enhance the image data generated in the Czochralski crystal growth process, improve the brightness, remove the halo and noise, and improve the image contrast, so as to improve the image quality, so that the processed images can better assist manual observation of the crystal growth process.

[0049] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for processing images of Czochralski crystal growth. Features: The process includes the following steps which are executed in sequence: S1: Collecting raw images during crystal growth ; S2: The original image The first layer sub-image is split into 4 * 4 , the first layer sub-image The second layer sub-image is divided into 8*8 , the second layer sub-image Split into 16 * 16 third layer sub-images , the Zero-DCE algorithm is used to extract the third layer sub-image Process and stitch the four adjacent sub-images together to get the second sub-image of 8 * 8 , the Zero-DCE algorithm is used to extract the second sub-image Process and stitch the four adjacent sub-images together to get the first sub-image of 4 * 4 , use Zero-DCE algorithm to extract the first sub-image Processing to get a new image ; S3: Segment the new image according to a preset segmentation form Perform the window moving operation and splitting operation three times in sequence to obtain a new third-layer sub-image , the Zero-DCE algorithm is used to generate the new third-layer sub-image After processing, a splicing operation is performed according to the preset segmentation form to obtain a new second layer sub-image , recursively operate until a new first image is obtained ; S4: Solving the first image Dark channel , from the dark channel Select a preset ratio of points from the image and correspond to the first image To find the mean of these points, as the first image The global estimate of , calculate the transmittance in the single crystal furnace according to the temperature and pressure in the single crystal furnace , the dark channel prior algorithm is used to calculate the observed image ; S5: Adaptive gamma for observed images Perform brightness enhancement processing to transform the observed image Convert to Space, extract observation image Brightness information ,Keep and The channel remains unchanged, and the brightness information To enhance, the formula for brightness information enhancement is as follows: ; In the formula, is the image brightness information after gamma transformation enhancement, is any pixel in the image, is the gamma correction factor, The expression is as follows: ; In the formula, is the maximum value of the brightness channel; is the minimum value of the brightness channel; S6: Output the image processed in step S5.

2. A method for processing a Czochralski crystal growth image as claimed in claim 1, Features: In step S1, an industrial camera with an ND filter on the lens is used to obtain the original image. Collection of.

3. A system for processing images of Czochralski crystal growth, comprising a memory and a processor, wherein the memory stores executable commands of the processor. Features: The processor is configured to implement the method according to any one of claims 1 or 2 by executing the executable instructions.

4. A storage medium having a computer program stored thereon, Features: When the computer program is executed, the method according to any one of claims 1 or 2 is implemented.