# US 2021/0279905 A1 · Crystal line growing state detection method, apparatus and device for silicon rod

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0279905 A1
- Document ID: US20210279905A1
- Status: Published application
- Kind: A1
- Application number: 17/256,494
- Filed: 2019-11-11
- Published: 2021-09-09
- Earliest claimed: 2018-11-26
- Assignees: Longi Green Energy Technology Co., Ltd.
- Applicants: Longi Green Energy Technology Co., Ltd.
- Inventors: LI GUO; QIAO LI; ZHANJUN XU
- Web page: https://patentgrep.com/patent/US20210279905A1

## CPC

- C30B 15/20: Controlling or regulating (controlling or regulating in general G05)
- G06T 7/62: of area, perimeter, diameter or volume
- C30B 15/26: using television detectors; using photo or X-ray detectors
- C30B 29/06: Silicon
- G06T 7/001: {using an image reference approach}
- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06T 2207/20072: Graph-based image processing
- G06T 2207/30242: Counting objects in image
- C30B: SINGLE-CRYSTAL GROWTH (by using ultra-high pressure, e.g. for the formation of diamonds, B01J3/06); UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL (zone-refining of metals or alloys C22B); PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE (casting of metals, casting of other substances by the same processes or devices B22D; working of plastics B29; modifying the physical structure of metals or alloys C21D, C22F); SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE (for producing semiconductor devices or parts thereof H10); APPARATUS THEREFOR
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method, apparatus and device for detecting a growth state of a crystalline line of a silicon rod, relating to the technical field of monocrystalline silicon, including in a process of constant-diameter growth of the silicon rod, acquiring a sample image of the silicon rod providing a detection area in the sample image, wherein the detection area overlaps with a crystalline-line growth line of the silicon rod, generating a grayscale-value curve of the detection area, and according to the grayscale-value curve of the detection area, determining the growth state of the crystalline line of the silicon rod on the crystalline-line growth line. The method alleviates the affection on the detection of crystalline lines by the fluctuation of the diameter of the silicon rod and the unclarity of the features of the crystalline lines, thereby improving the accuracy and the efficiency of the detection on the crystalline lines.

## Claims

### Claim 1

A method for detecting a growth state of a crystalline line of a silicon rod, wherein the method comprises:
in a process of constant-diameter growth of the silicon rod, acquiring a sample image of the silicon rod;
providing a detection area in the sample image, wherein the detection area overlaps with a crystalline-line growth line of the silicon rod, and in the process of the constant-diameter growth of the silicon rod, the crystalline line of the silicon rod grows along the crystalline-line growth line;
generating a grayscale-value curve of the detection area; and
according to the grayscale-value curve, determining the growth state of the crystalline line of the silicon rod on the crystalline-line growth line.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the detection area comprises: a line segment perpendicular to an axial direction of the silicon rod, or a rectangular area perpendicular to an axial direction of the silicon rod, wherein a plane of the rectangular area is perpendicular to the axial direction of the silicon rod.

### Claim 3 (depends on 2)

The method according to claim 2, wherein if the detection area is the line segment perpendicular to the axial direction of the silicon rod, the step of generating the grayscale-value curve of the detection area comprises:
by starting from one end of the line segment, generating a grayscale-value curve corresponding to the line segment.

### Claim 4 (depends on 2)

The method according to claim 2, wherein if the detection area is the rectangular area perpendicular to the axial direction of the silicon rod, the step of generating the grayscale-value curve of the detection area comprises:
by starting from one end of the rectangular area, dividing the rectangular area into a plurality of identical sub-areas;
calculating an average grayscale value of each of the sub-areas; and
according to the average grayscale values of all of the sub-areas, generating a grayscale-value curve corresponding to the rectangular area.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the step of, according to the grayscale-value curve, determining the growth state of the crystalline line of the silicon rod on the crystalline-line growth line comprises:
according to the grayscale-value curve, calculating a variance value of the grayscale values in the grayscale-value curve;
if the variance value of the grayscale values is greater than or equal to a variance-value threshold, determining the growth state to be a continued state; and
if the variance value of the grayscale values is less than the variance-value threshold, determining the growth state to be a line-breaking state.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the step of, according to the grayscale-value curve, determining the growth state of the crystalline line of the silicon rod on the crystalline-line growth line comprises:
if the grayscale-value curve has a crystalline-line characteristic peak, determining the growth state to be a continued state; and
if the grayscale-value curve does not have the crystalline-line characteristic peak, determining the growth state to be a line-breaking state.

### Claim 7 (depends on 1)

The method according to claim 1, wherein after the step of acquiring the sample image of the silicon rod, the method further comprises:
according to a predetermined image-enhancement algorithm, performing image-enhancement processing to the sample image.

### Claim 8

(canceled)

### Claim 9

(canceled)

### Claim 10

(canceled)

### Claim 11

A method for detecting a broken line, wherein the method comprises:
processing a current frame of an acquired growth image of a single crystal and a previous one frame of the growth image, and obtaining a quantity of moving pixels in the current frame of the growth image;
within a unit time, counting up to obtain a quantity of moving pixels in all of frames of the growth image and information of a single-crystal-growth duration; and
according to the information, determining whether the single crystal has a broken line.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the step of processing each of the image frames, and obtaining the quantity of moving pixels in each of the image frames comprises:
selecting an image measurement area;
performing data processing to each of pixels within the image measurement area, and obtaining a velocity amplitude of each of the pixels;
if the velocity amplitude of a pixel is greater than a preset threshold, determining the pixel to be a moving pixel; and
counting up a quantity of the moving pixels within the image measurement area.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the step of selecting the image measurement area comprises:
according to a position of a light ring, selecting an image measurement area in each of the image frames.

### Claim 14 (depends on 13)

The method according to claim 13, wherein the step of selecting the image measurement area comprises:
according to a grayscale value, selecting an image measurement area in each of the image frames, wherein a maximum grayscale value is 200.

### Claim 15 (depends on 13)

The method according to claim 13, wherein the step of performing data processing to each of the pixels within the image measurement area, and obtaining the velocity amplitude of each of the pixels comprises:
acquiring a velocity in a first direction and a velocity in a second direction of an i-th pixel within the image measurement area; and
according to the velocity in the first direction, the velocity in the second direction and a first formula, obtaining a velocity amplitude of the i-th pixel within the image measurement area, wherein the first formula comprises:
{
m
i
=
u
2
+
v
2
*
f
⁡
(
θ
i
)
,
0
⁢
°
≤
θ
i
≤
360
⁢
°
,
and
⁢
⁢
θ
i
≠
90
⁢
°
m
i
=
u
2
+
v
2
,
θ
i
=
90
⁢
°
wherein i=1, 2, 3, . . . n, mi represents the velocity amplitude of the i-th pixel within the image measurement area, θi represents an included angle between the first direction and the second direction, f(θi) represents a function relation formula that is relevant to the θi, ui represents the velocity in the first direction, and vi represents the velocity in the second direction.

### Claim 16 (depends on 15)

The method according to claim 15, wherein the step of acquiring the velocity in the first direction and the velocity in the second direction of the pixel of any point within the image measurement area comprises:
according to a coordinate-point position of the i-th pixel within the image measurement area, an image-collection time interval and a second formula, obtaining the velocity in the first direction and the velocity in the second direction of the i-th pixel within the image measurement area, wherein the second formula is: (ui, vi)=[(xi2, yi2)−(xi1, yi1)]/t ;
wherein x represents the first direction, y represents the second direction, (xi1, yi1) represents a coordinate-point position of the i-th pixel within the image measurement area at the (s-1)-th frame, (xi2, yi2) represents a coordinate-point position of the i-th pixel within the image measurement area at the s-th frame, and t represents the image-collection time interval between the s-th frame and the (s-1)-th frame.

### Claim 17 (depends on 16)

The method according to claim 16, wherein the preset threshold is obtained according to the velocity amplitudes of all of the pixels within the image measurement area, by the following step:
according to the velocity amplitudes of all of the pixels within the image measurement area, a quantity of the pixels within the image measurement area and a third formula, obtaining the preset threshold, wherein the third formula is:
Y
=
∑
1
k
⁢
M
g
/
N
;
wherein Y represents the preset threshold, Mg represents a velocity amplitude of the g-th pixel within a coverage of the light ring, g=1, 2, 3, . . . k, and N represents a quantity of the pixels within the coverage of the light ring.

### Claim 18 (depends on 11)

The method according to claim 11, wherein the step of, according to the information, determining whether the single crystal has a broken line comprises:
according to the quantity of the moving pixels in all of the frames of the growth image and a waveform-variation curve of the single-crystal-growth duration, if a duration of an interval between two neighboring peaks in the waveform-variation curve is greater than a preset duration, determining that the single crystal has a broken line.

### Claim 19 (depends on 11)

A device for detecting a broken line, wherein the device comprises a second processor and a second memory, the second memory stores at least one instruction, and the instruction is loaded and executed by the second processor to implement the steps of the method for detecting a broken line according claim 11.

### Claim 20

(canceled)