Computer-implemented method for defect analysis, apparatus for defect analysis, computer-program product, and intelligent defect analysis system

Patent metadata

Field	Value
Publication number	US-11972548-B2
Application number	US-202017438734-A
Country	US
Kind code	B2
Filing date	Dec 3, 2020
Priority date	Dec 3, 2020
Publication date	Apr 30, 2024
Grant date	Apr 30, 2024

Abstract

A computer-implemented method for defect analysis is provided. The computer-implemented method includes obtaining a plurality of sets of defect point coordinates, a respective set of the plurality of sets of detect point coordinates including coordinates of defect points in a respective substrate of a plurality of substrates, the coordinates of defect points in the respective substrate being coordinates in an image coordinate system; combining the plurality of sets of defect point coordinates according to the image coordinate system into a composite set of coordinates to generate a composite image; and performing a clustering analysis to classify defect points in the composite set in the composite image into a plurality of clusters.

First claim

What is claimed is:

1. A computer-implemented method for defect analysis, comprising:
obtaining a plurality of sets of defect point coordinates, a respective set of the plurality of sets of defect point coordinates comprising coordinates of defect points in a respective substrate of a plurality of substrates, the coordinates of defect points in the respective substrate being coordinates in an image coordinate system;
combining the plurality of sets of defect point coordinates according to the image coordinate system into a composite set of coordinates to generate a composite image;
performing a clustering analysis to classify defect points in the composite set in the composite image into a plurality of clusters
determining a plurality of contours respectively of at least a plurality of selected clusters of the plurality of clusters, a respective one of the plurality of contours comprising a plurality of edge defect points in a respective one of the plurality of selected clusters;
applying a fitting algorithm to edge defect points of the plurality of selected clusters to generate a plurality of mask areas respectively corresponding to the plurality of selected clusters; and
generating a plurality of feature vectors respectively of the plurality of mask areas.

2. The computer-implemented method of claim 1 , further comprising obtaining a plurality of selected clusters from the plurality of clusters;
wherein a number of defect points in each of the plurality of selected clusters is greater than a threshold number.

3. The computer-implemented method of claim 1 , wherein generating the plurality of feature vectors comprises:
generating Hu geometric moment m i,j and center-to-center distance M i,j of a respective one of the plurality of mask areas, wherein m i,j =Σ (x,y)∈A x i y j ;
calculating defect point density p, area a, center of mass O (O x , O y ), and direction θ of the respective one of the plurality of mask areas; and
generating a respective one of the plurality of feature vectors for the respective one of the plurality of mask areas.

4. The computer-implemented method of claim 3 , wherein a respective one of the plurality of feature vectors is expressed as:
F=[ρ,a,Ox,Oy,θ,L,W,r] T ;
wherein

ρ
=

N
a

,

 N stands tor a number of defect points in the respective one of the plurality of mask areas, a stands for an area of the respective one of the plurality of mask areas;

O
x

=

m
10

a

;

⁢

O
y

=

m
01

a

;

⁢

θ
=

-
0.5

⁢
atc
⁢

tan
⁡
(

2
⁢

M
11

M
02

-

M
20

)

;

⁢

M
ij

=

∑

(

x
,
y

)

∈
𝒜

(

x
-

O
x

)

i

⁢

(

y
-

O
y

)

j

;

⁢

r
=

L
W

;;

L stands for a length of a minimal external rectangle of the respective one of the plurality of mask areas; and
W stands for a width of a minimal external rectangle of the respective one of the plurality of mask areas.

5. The computer-implemented method of claim 1 , wherein the plurality of contours are determined using an alpha shapes-based method.

6. The computer-implemented method of claim 1 , further comprising:
assigning one or more selected mask areas of the plurality of mask areas as a plurality of defect aggregation areas;
wherein feature vectors respectively of the one or more selected mask areas satisfy a threshold condition.

7. The computer-implemented method of claim 6 , further comprising:
comparing parameters of first defect points inside the one or more selected mask a

Assignees

• Boe Technology Group Co Ltd

Inventors

• Wang Haijin
• Zeng Jianfeng

Classifications

• G06T7/0002 Primary

  Inspection of images, e.g. flaw detection · CPC title

• G06T7/50

  Depth or shape recovery · CPC title

• G06V10/7625

  Hierarchical techniques, i.e. dividing or merging patterns to obtain a tree-like representation; Dendograms · CPC title

• G06V10/80

  Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level (multimodal speaker identification or verification G10L17/10) · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title