結言 - むら欠陥の官能検査の自動化に関する検討 - 光学機能シートの低コントラスト欠陥検出に関する研究

第７章むら欠陥の官能検査の自動化に関する検討

7.8 結言

ディスプレイの部材として用いられる反射防止膜および格子構造フィルムに発生するむら欠陥について，以下の検討を行った．

(1) 欠陥の特徴量の度数分布に基づいて目視官能検査と同等の合否判定を行う手

法を確立した．反射防止膜および格子構造フィルムの各欠陥に適用し，効果を検証した．

(2) 格子構造フィルムのむら欠陥発生要因が格子の線幅の不均一によることに着

目し，レーザの回折パターンを利用した欠陥検査方法を提案した．

(3) 各種欠陥モデルを使用して回折パターンをシミュレーションにより求めた．

その結果，欠陥の存在により発生するパターンの変化を明らかにした．

(4) 実際の欠陥サンプルに本手法を適用し，正常品によるパターンとの識別が可

能であることを確認した．

本手法では，むらを表す特徴量とその頻度というわずかな情報に集約したにも関わらず目視官能検査と良く一致する合否判定基準を確立することができた．また，本論文中では取扱う特徴量を回折パターンのスポットの重心あるいは等価膜厚としたが，これに限るものではなく一般的な濃淡差や色度差を特徴量としてもよい．そのため，本手法はむら検査全般に適用できる可能性が高いと考えられる．

参考文献

[1] たとえば，松田隆夫：知覚心理学の基礎, 培風館, (2000) 23-101.

[2] F.C.Volkmann: Human visual suppression, Vision Research,26,9, (1986) 1401-1416.

[3] B.L.Zuber: Saccadic suppression, Quartery progress report, I.T., 75, (1964) 190-191.

[4] H.P.Kleinknecht and H.Meier: Linewidth measurement on IC masks and wafers by grating test patterns, Appl. Opt. 19, 4, (1980) 525-533.

[5] 高谷裕浩, 三好隆志, 外山潔, 斎藤勝政：回折パターンによる極微細溝形状の測定評価に関する研究, 精密工学会誌, 57, 11, (1991) 2041-2047.

[6] L.S.Watkins: Inspestion of Integrated Circuit Photomasks with Intensity Spatial Filters, Proc. IEEE, 57, 9, (1969) 1634-1639.

[7] P.M.Will and K.S.Pennington: Filtering of Defects in Integrated Circuits with Orientation Independence, Appl. Opt., 10, 9, (1971) 2097-2100.

[8] A.Iwamoto and H.Sekizawa: Defect-type discriminationg optical system, Apll. Opt., 20, 9, (1981) 1724-1726.

[9] 深谷次助, 佐々木彰：スリット開口の Fraunhofer回折による旋削面粗さ測定, 精密機械, 51, 4, (1985) 173-178.

[10] C.Ciamberlini, F.Francini, G.Longobardi, P.Sansoni and B.Tiribilli: Defect detection in textured materials by optical filtering with structured detectors and self-adaptable masks, Opt. Eng., 35, 3, (1996) 838-844.

[11] Seung-Woo Kim, Sang-Yoon Lee and Dong-Seon Yoon: Rapid pattern inspection of shadow masks by machine vision integrated with Fourier optics, Opt. Eng., 36, 12, (1997) 3309-3311.

[12] J.W.Goodman: Introduction to Fourier Optics Second Eition, McGRAW-HILL, (1996).

[13] 飯塚啓吾：光工学, 共立出版, (1977) 29-62.

Fig.7.1 Feature and Frequency Method(FF-method)

OK (Non-defective sample) NG (Defective sample)

50mm (500pixels)

(a) Observed images of anti-reflection coating samples

79.0 79.5 80.0 80.5 81.0

0 100 200 300 400 500

Position x

pixel E qui va le n t T h ic kne ss nm

NG (Defective sample) OK (Non-defective sample)

(b) Comparison of equivalent thickness between defective and non-defective sample

(b) Comparison of frequency distribution of equivalent thickness 0

50 100 150 200 250

0.0 0.5 1.0 1.5 2.0 2.5

Fluctuation of equivalent thickness nm

Frequency

NG (Defective sample) OK (Non-defective sample) Borderline of non-defective film

Total (Grade of defect) [

Comparative evaluation sheet

]

Date : 2002.4.24

Object : Color shading defect (CG) Operator : Chama

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ＣＧ＃

1 0 1 1 1 1 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 10 OK 2 1 0 1 1 0 0 0 0 0 0 1 1 1 0 1 0 0 0 0 1 1 1 0 0 0 0 0 2 12 OK 3 1 1 0 1 0 0 0 0 0 0 1 1 1 1 0 1 0 0 0 1 1 1 1 0 0 0 0 0 12 OK 4 1 1 1 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 0 1 0 0 0 10 OK 5 1 2 2 2 0 1 1 1 1 0 1 2 2 1 0 1 0 0 0 1 1 2 0 1 0 0 0 0 23 NG 6 2 2 2 2 1 0 1 1 0 0 2 2 2 1 0 0 0 0 0 2 2 2 1 0 0 0 0 0 25 NG 7 2 2 2 2 1 1 0 1 1 0 0 2 1 1 1 1 0 1 0 1 2 1 1 1 0 0 0 0 25 NG 8 2 2 2 2 1 1 1 0 1 1 2 2 2 1 1 1 0 0 0 2 2 2 1 0 0 0 0 0 29 NG 9 2 2 2 2 1 2 1 1 0 1 2 2 2 1 1 1 0 0 0 2 2 2 2 1 2 0 0 0 34 NG 10 2 2 2 2 2 2 2 1 1 0 2 2 0 2 1 1 1 1 1 2 2 2 2 2 1 0 0 0 38 NG 11 1 1 1 1 1 0 0 0 0 0 0 1 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 12 OK 12 1 1 1 1 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 1 1 1 0 0 0 0 0 11 OK 13 1 1 1 1 0 0 1 0 0 0 1 1 0 1 0 0 0 0 0 1 1 1 0 0 0 0 0 0 11 OK 14 2 2 1 2 1 1 1 1 1 0 1 2 1 0 1 1 0 0 0 1 2 1 1 0 0 0 0 0 23 NG 15 2 1 2 2 2 2 1 1 1 1 1 2 2 1 0 1 1 0 0 2 1 2 1 1 0 1 1 0 32 NG 16 2 2 1 2 1 2 1 1 1 1 2 1 2 1 1 0 0 1 1 2 2 2 1 1 1 0 0 0 32 NG 17 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 2 0 1 1 2 2 2 2 1 2 1 1 0 45 NG 18 2 2 2 2 2 2 1 2 2 1 2 2 2 2 2 1 1 0 1 2 2 2 2 1 1 1 1 1 44 NG 19 0 2 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1 1 0 2 2 2 2 1 1 1 1 1 43 NG

Fig.7.2 An example of a pair of CG images

Fig.7.4 Configuration of the inspection borderline

(a) Experimental method

Plots show the histograms involving OK samples, which are scored less than 20 points.

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 Fluctuation of equivalent thickness nm

Frequency

Borderline

(b) Method by distribution functions

Plots by solid lines show the Weibull distribution curves with different coefficient α.

The borderline is configured to cover the distribution curve involving the limit sample.

0.00 0.50 1.00 1.50 2.00 2.50

0.2 0.3 0.5 0.7 coefficient α Borderline

Fluctuation of equivalent thickness nm

Frequency Weibull distribution :

⎟⎟⎠

⎞

⎜⎜⎝

⎛−

= ⁻

α α

m x

x mx

f( ) exp

(m=3)

Visual inspection point : 10 FF method inspection : OK

M313

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 10

Fluctuation of equivalent thickness

Frequency

Visual inspection point : 11 FF method inspection : OK

(b) Sample CG #13

M213

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 11

Fluctuation of equivalent thickness

Frequency

(a) Sample CG #3 Visual inspection point : 12

FF method inspection : OK

M112

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 12

Fluctuation of equivalent thickness

Frequency

M132

100 150 200 250 300 350 400

Frequency

Fig.7.5 Comparison between FF-method and visual inspection by human operator (continued)

Visual inspection point : 45 FF method inspection : NG

(e) Sample CG #17

M231

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 45

Fluctuation of equivalent thickness

Frequency

Visual inspection point : 27 FF method inspection : NG

(f) Sample CG #23

M321

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 27

Fluctuation of equivalent thickness

Frequency

Visual inspection point : 25 FF method inspection : NG

(g) Sample CG #6

M122

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 25

Fluctuation of equivalent thickness

Frequency

(h) Sample CG #28 Visual inspection point : 49

FF method inspection : NG

M333

0 50 100 150 200 250 300 350 400

0.00 0.50 1.00 1.50 2.00 2.50 49

Fluctuation of equivalent thickness

Frequency

Fig.7.6 Grid structure film and shading defect

Grid structure film 1 mm

Shading defect

(a) Configuration of grid structure film and shading defect

(b) Defect model

(x0,y0)

(xi,yi) yi

z (Optical axis) Electric field

g(x0,y0)

z=0

Electric field u(xi,yi) r = zi

2+ (x0-xi)² + (y0-yi)²

(a) Configuration of grid structure

g(x0)

b c

(b) Intensity profile

Fig.7.8 Electric field on grid structure

Fig.7.9 Power profile of Fourier transform of electric field on grid structure

0 . 0 0 0 . 0 5 0 . 1 0 0 . 1 5 0 . 2 0

0 1 0 2 0 3 0

p o sitio n xi [m m ]

Normalized power

Envelope curve (sinc²(afx)) Power profile P(xi) = |G(xi)| ²

Normalized power

Position xi mm

1mm 1mm

Fig.7.10 An example model of electric field on grid structure film (computer-generated image)

Beam characteristic : Gaussian beam (width w = 0.4 mm, in phase) Grid structure : line width a = 20μm , grid interval b = 195μ m

(a) Simulation

(b) Experimental

alized power S im ulation

Experim ental result

Fig.7.12 Variations of envelop curve of diffraction pattern with the unevenness of grid line width

0 10 20 30 40

Position x

[mm]

N o rm al ized p o w er

^(a)

(b) (c)

thick

thin

thick thin (a) (b) (c)

varied fixed (20μm) (a) a=25μm in left side

(b) uniform grid (a=20μm) (c) a=15μm in left side

Position xi mm

Normalized power

0 5 P osition x_i [m m ]10 15

Normalized power

195μm interval (ordinary) 175μm

interval (extraordinary)

Normalized power

0 5 10 15 Position ximm

Fig.7.13 Diffraction pattern with the grid interval unevenness

Extraordinat spots are observed between ordinary spots.

Order of spots

0 7 0 8 0 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8

(b) Diffraction pattern of line width defective film (a) Diffraction pattern of normal grid structure film Main lobe First side lobe

He-Ne laser

Grid structure film

z0=0

zi=500 mm Diffraction

pattern Screen

CCD area sensor conveyer

Fig.7.14 Configuration of experimental system

Fig.7.16 Variations of diffraction pattern with laser incident position on the grid structure film

Each pattern is observed under the same film with different incident position of laser beam.

Fig.7.17 Variations of the center of diffraction pattern involving a normal film and defective films

The longitudinal axis stands for the center position of the spots which are comprised of the first side lobe of the diffraction pattern.

13 15 17 19

0 50 100 150 200 250 300

Position x⁰ [mm]

Position xi

NG6_UPPER NG7_CENTER

OK7_UPPER OK6_CENTER

(Numerical order of spots)

(a) Fatal defective film (b) Defective film (c) Limit sample film (d) Normal film

(b) (c) (a) (d)

0.0 0.5 1.0 1.5 2.0 2.5

0 50 100 150 200 250 300

Pos ition x⁰ [mm]

Fluctuation of center

NG6_UPPER NG7_CENTER

OK7_UPPER OK6_CENTER

(a) Fatal defective film (b) Defective film (c) Limit sample film (d) Normal film

Fig.7.18 Fluctuation of center positions of diffraction pattern The longitudinal axis stands for the maximum minus the minimum values of Fig.7.11 within 10 mm width.

Fig.7.19 Frequency distribution of fluctuation

50 100 150 200

0 10 20 30 40

Fluctuation band

Frequency

NG6_UPPER NG7_CENTER

OK7_UPPER OK6_CENTER

Border line of normal and defective film

Scale of fluctuation of center

(a) Fatal defective film (b) Defective film (c) Limit sample film (d) Normal film

(d) (c) (b)

(a)

ドキュメント内光学機能シートの低コントラスト欠陥検出に関する研究 (ページ 121-135)

結言

第７章 むら欠陥の官能検査の自動化に関する検討

7.8 結言

(1) 欠陥の特徴量の度数分布に基づいて目視官能検査と同等の合否判定を行う手

(2) 格子構造フィルムのむら欠陥発生要因が格子の線幅の不均一によることに着

(3) 各種欠陥モデルを使用して回折パターンをシミュレーションにより求めた．

(4) 実際の欠陥サンプルに本手法を適用し，正常品によるパターンとの識別が可

参 考 文 献

Fig.7.1 Feature and Frequency Method(FF-method)

79.0 79.5 80.0 80.5 81.0

0 100 200 300 400 500

Position x

pixel E qui va le n t T h ic kne ss nm

NG (Defective sample) OK (Non-defective sample)

Comparative evaluation sheet

Fig.7.2 An example of a pair of CG images

Fig.7.4 Configuration of the inspection borderline

α α

Fig.7.5 Comparison between FF-method and visual inspection by human operator (continued)

Fig.7.6 Grid structure film and shading defect

(a) Configuration of grid structure film and shading defect

Fig.7.8 Electric field on grid structure

Fig.7.9 Power profile of Fourier transform of electric field on grid structure

Fig.7.10 An example model of electric field on grid structure film (computer-generated image)

Beam characteristic : Gaussian beam (width w = 0.4 mm, in phase) Grid structure : line width a = 20μm , grid interval b = 195μ m

Fig.7.12 Variations of envelop curve of diffraction pattern with the unevenness of grid line width

0 10 20 30 40

Position x

[mm]

N o rm al ized p o w er

Fig.7.13 Diffraction pattern with the grid interval unevenness

Extraordinat spots are observed between ordinary spots.

Fig.7.14 Configuration of experimental system

Fig.7.16 Variations of diffraction pattern with laser incident position on the grid structure film

Each pattern is observed under the same film with different incident position of laser beam.

Fig.7.17 Variations of the center of diffraction pattern involving a normal film and defective films

The longitudinal axis stands for the center position of the spots which are comprised of the first side lobe of the diffraction pattern.

Fig.7.18 Fluctuation of center positions of diffraction pattern The longitudinal axis stands for the maximum minus the minimum values of Fig.7.11 within 10 mm width.

Fig.7.19 Frequency distribution of fluctuation

第７章むら欠陥の官能検査の自動化に関する検討

参考文献