JPH09257642A - Glass substrate defect type determination method - Google Patents

Glass substrate defect type determination method

Info

Publication number
JPH09257642A
JPH09257642A JP8087049A JP8704996A JPH09257642A JP H09257642 A JPH09257642 A JP H09257642A JP 8087049 A JP8087049 A JP 8087049A JP 8704996 A JP8704996 A JP 8704996A JP H09257642 A JPH09257642 A JP H09257642A
Authority
JP
Japan
Prior art keywords
defect
glass substrate
substrate
light
scratch
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP8087049A
Other languages
Japanese (ja)
Inventor
Masaomi Takeda
正臣 武田
Kenji Aiko
健二 愛甲
Noboru Kato
昇 加藤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi High Tech Corp
Original Assignee
Hitachi Electronics Engineering Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Electronics Engineering Co Ltd filed Critical Hitachi Electronics Engineering Co Ltd
Priority to JP8087049A priority Critical patent/JPH09257642A/en
Publication of JPH09257642A publication Critical patent/JPH09257642A/en
Pending legal-status Critical Current

Links

Landscapes

  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Testing Of Optical Devices Or Fibers (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

(57)【要約】 【課題】 ガラス基板に存在する3種の欠陥を検出し、
その種別が異物、スクラッチ傷、気泡のいずれであるかを
判定する。 【解決手段】 基板1の裏面側より、波長帯域の異なる
(例えば白色光束と赤色光束)を、基板1に対して約2
0°の入射角θ1 と約60°の入射角θ2 でそれぞれ投
射し、基板の表面側に、光軸が基板に垂直な受光系を設
け、その集光レンズにより各欠陥の散乱光のみを集光
し、ハーフミラーHMとフィルタFa,Fb により、散乱
光を白色光と赤色光に分割して、それぞれをCCDセン
サ(A),(B)で受光し、両CCDセンサの出力する欠
陥信号Sa,Sb の強度をia,ib 、欠陥信号Sa または
b を画像処理して求めた各欠陥の長さと幅をL,Wと
し、両強度の比ia /ib と、L,Wの比L/Wとを、
判定基準R1,R2 にそれぞれ比較して、異物Ki、スクラ
ッチ傷Ks、気泡Kb を判定する。
(57) 【Abstract】 PROBLEM TO BE SOLVED: To detect three kinds of defects existing in a glass substrate,
It is determined whether the type is a foreign substance, a scratch, or a bubble. SOLUTION: A wavelength band (for example, a white light flux and a red light flux) having different wavelength bands from the back side of the substrate 1 is about 2 with respect to the substrate 1.
Projecting respectively 0 ° angle of incidence theta 1 and about 60 ° of the incident angle theta 2, the surface side of the substrate, providing a vertical light receiving system optical axis in the substrate, only the scattered light of each defect by the condenser lens Is collected, the scattered light is divided into white light and red light by the half mirror HM and the filters F a and F b , and the CCD sensors (A) and (B) receive each, and the output of both CCD sensors The intensities of the defect signals S a and S b are i a and i b , and the length and width of each defect obtained by image processing the defect signal S a or S b are L and W, and the ratio of both intensities i a / i b and L / W ratio L / W are
The foreign matters K i , the scratches K s , and the bubbles K b are compared with the determination criteria R 1 and R 2 , respectively.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【発明の属する技術分野】この発明は、ガラス基板の欠
陥種別の判定方法に関し、詳しくは液晶パネル用のガラ
ス基板の表面、裏面または内部に存在する3種類の欠陥
の、種別を判定する方法に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of determining a defect type of a glass substrate, and more particularly to a method of determining the type of three types of defects existing on the front surface, back surface or inside of a glass substrate for a liquid crystal panel. .

【0002】[0002]

【従来の技術】液晶パネルを構成するTFT基板は、ガ
ラス基板を素材としてその表面に多数のTFT素子が形
成される。この表面に傷などの欠陥が存在すると、その
部分にはTFT素子が良好に形成されず、従ってTFT
基板の品質を阻害するので、欠陥検査装置により欠陥の
有無を検査して所要の措置がとられている。欠陥検査装
置においては、ガラス基板に対してレーザビームを投射
し、欠陥の散乱光を受光器に受光して欠陥を検出し、そ
の欠陥信号より欠陥のサイズを算出し、これに欠陥の存
在する座標値を付加した欠陥データが出力されている。
2. Description of the Related Art A TFT substrate that constitutes a liquid crystal panel is made of a glass substrate, and a large number of TFT elements are formed on the surface thereof. If there are defects such as scratches on this surface, the TFT element is not formed well in that portion, and therefore the TFT
Since it impairs the quality of the substrate, the defect inspection apparatus inspects for the presence of defects and takes necessary measures. In the defect inspection device, a laser beam is projected onto the glass substrate, the scattered light of the defect is received by a photodetector, the defect is detected, the size of the defect is calculated from the defect signal, and the defect is present in this. The defect data with the coordinate value added is output.

【0003】[0003]

【発明が解決しようとする課題】さて、ガラス基板(以
下単に基板という)の欠陥には各種のものがあり、代表
的なものには、基板の表面または裏面に存在する異物、
スクラッチ傷、および気泡の3種がある。ただし異物に
は、塵埃などが付着した付着異物と基板の製造過程で固
着した固着異物がある。スクラッチ傷は両面の研磨や基
板の取り扱い中に生じ、気泡は製造過程で基板の内部に
発生する。これらのうち、付着異物は再洗浄により、固
着異物は再研磨によりそれぞれ除去可能であるが、スク
ラッチ傷と気泡はこれができない。従って基板の製造と
管理上、各欠陥の種別を知ることが必要である。しかし
ながら上記の欠陥検査装置では、これらは区別されずに
単に欠陥として検出されている。そこで欠陥の種別を判
定する有効な手段が要請されている。この発明は、ガラ
ス基板に存在する3種の欠陥を一括して検出し、その種
別が、異物、スクラッチ傷、気泡のいずれであるかを判
定することが課題である。
There are various types of defects in a glass substrate (hereinafter simply referred to as a substrate), and typical ones are foreign substances existing on the front surface or the back surface of the substrate,
There are three types of scratches and bubbles. However, the foreign matter includes an adhering foreign matter to which dust or the like adheres and a fixed foreign matter to which the foreign matter adheres in the process of manufacturing the substrate. Scratch scratches occur during double-side polishing and handling of the substrate, and bubbles are generated inside the substrate during the manufacturing process. Of these, the adhered foreign matter can be removed by re-cleaning, and the adhered foreign matter can be removed by re-polishing, but scratches and air bubbles cannot. Therefore, it is necessary to know the type of each defect in manufacturing and managing the substrate. However, in the above-described defect inspection apparatus, these are not distinguished but are simply detected as defects. Therefore, effective means for determining the type of defect is required. An object of the present invention is to collectively detect three types of defects existing on a glass substrate and determine whether the type is a foreign substance, a scratch, or a bubble.

【0004】[0004]

【課題を解決するための手段】この発明は、上記の課題
を解決したガラス基板の欠陥種別判定方法であって、基
板の裏面側に設けた投光系により、波長帯域の異なる少
なくとも2本の光束を、基板の検査位置に対して小さい
入射角と大きい入射角でそれぞれ投射する。基板の表面
側に、光軸が基板に垂直な受光系を設け、受光系が有す
る各光束が入射しない狭い受光視角の集光レンズによ
り、各欠陥による各光束の散乱光を集光し、集光された
散乱光を上記の異なる波長帯域に分割して、それぞれの
CCDイメージセンサにより受光する。両CCDイメー
ジセンサの出力する欠陥信号Sa,Sb の強度ia,ib
比ia /ib と、欠陥信号Sa またはSb を画像処理し
て求めた各欠陥の長さLと幅Wの比L/Wを算出し、両
比ia /ib ,L/Wに対する判定基準をR1,R2 とし
て、下記の各判定条件: (ia /ib >R1 ,L/W>R2 ):Ks ………条件(1) (ia /ib >R1 ,L/W≦R2 ):Kb ………条件(2) (ia /ib ≦R1 ,L/W>R2 ):Ks ………条件(3) (ia /ib ≦R1 ,L/W≦R2 ):Ki ………条件(4) により、異物Ki 、スクラッチ傷Ks 、気泡Kb をそれ
ぞれ判定する。
SUMMARY OF THE INVENTION The present invention is a defect type determination method for a glass substrate which solves the above-mentioned problems, wherein at least two different wavelength bands are provided by a light projecting system provided on the back side of the substrate. The light beam is projected at a small incident angle and a large incident angle with respect to the inspection position of the substrate. A light receiving system whose optical axis is perpendicular to the substrate is provided on the surface side of the substrate, and the light collecting system has a narrow light receiving viewing angle condensing lens that collects the scattered light of each light beam and collects it. The scattered light emitted is divided into the above-mentioned different wavelength bands and received by the respective CCD image sensors. Output defect signal S a of both the CCD image sensor, the intensity i a of S b, and the ratio i a / i b of i b, defect signal S a or the length of each defect found by the image processing S b L and calculating the ratio L / W of the width W, both the ratio i a / i b, the criteria for L / W as R 1, R 2, each determined under the following conditions: (i a / i b> R 1, L / W> R 2): K s ......... condition (1) (i a / i b> R 1, L / W ≦ R 2): K b ......... condition (2) (i a / i b ≦ R 1, L / W> R 2): K s ......... condition (3) (i a / i b ≦ R 1, L / W ≦ R 2): the K i ......... condition (4), The foreign matter K i , the scratch K s , and the bubble K b are determined.

【0005】[0005]

【発明の実施の形態】まず、図1により3種の欠陥の形
状と散乱光の指向性を説明し、図2と図3により、この
発明の欠陥種別判定方法の原理を説明する。図1におい
て、(a) は各欠陥の形状などを説明するもので、異物K
i は表面に凹凸のある粒状のものが多く、そのサイズは
まちまちで、TFT素子に比較してある大きさ以上のも
のが有害である。スクラッチ傷Ks は通常、基板1の表
面を溝状に削ったもので、その幅と深さは比較的に小さ
いが、これに比べてある程度長いものが有害とされる。
気泡Kb は通常、球形またはその変形で、そのサイズは
まちまちであるが、余程大きいものが有害とされる。
(b) は各欠陥の指向特性を説明するもので、各欠陥に光
束Lを投射するとそれぞれは形状に従った散乱光を散乱
する。異物Ki の散乱光は光束Lの投射方向にあまり関
係なく、上半球の方向にほぼ等しい強度で散乱し、すな
わち無指向性である。スクラッチ傷Ks の散乱光は、光
束Lの投射方向に大きく関係し、投射方向によっては、
図示のように垂直上方に強いピークをなし、低角度ほど
弱い指向性がある。気泡Kb の散乱光は異物Ki とスク
ラッチ傷Ks の中関的な指向性である。なお各欠陥の散
乱光の強度は、それぞれのサイズに依存して変化し、サ
イズが大きいほど大きい。
DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the shapes of three types of defects and the directivity of scattered light will be described with reference to FIG. 1, and the principle of the defect type determination method of the present invention will be described with reference to FIGS. In FIG. 1, (a) illustrates the shape of each defect, and the foreign matter K
Many of i have a granular shape with unevenness on the surface, and the sizes thereof are various, and those having a size larger than a certain size are harmful. The scratches K s are usually formed by grooving the surface of the substrate 1, and the width and depth thereof are relatively small, but the scratches K s longer than this are harmful.
The bubbles K b are usually spherical or a deformation thereof, and their sizes are different, but if they are too large, they are harmful.
(b) illustrates the directional characteristics of each defect. When the light flux L is projected on each defect, each scatters scattered light according to the shape. The scattered light of the foreign matter K i is scattered in the direction of the upper hemisphere with almost the same intensity regardless of the projection direction of the light flux L, that is, it is omnidirectional. Scattered light scratches K s is closely related to the projection direction of the light beam L, the projection direction,
As shown in the figure, there is a strong vertical peak, and the lower the angle, the weaker the directivity. The scattered light of the bubble K b has a direct relationship with the foreign matter K i and the scratch K s . The intensity of scattered light of each defect changes depending on its size, and the larger the size, the greater the intensity.

【0006】図2において、基板1の裏面側から小さい
入射角θ1 と大きい入射角θ2 をなして、波長帯域の異
なる、例えば白色光束と赤色光束をそれぞれ投射する。
基板1の表面に欠陥K、または裏面に欠陥K’がある
と、これらは散乱光を散乱する。表面側に、光軸が表面
に垂直をなす受光系を設け、その集光レンズの受光視角
θ3 を、基板1を透過した両光束が入射しないように狭
くして、欠陥KまたはK’の散乱光のみを集光し、これ
をハーフミラーHMにより分割し、透過した一方をフィ
ルタFA により赤色を除去し、反射した他方をフィルタ
B により赤色を透過して、それぞれをCCDイメージ
センサ(A),(B)により受光する。両CCD(A),
(B)が出力する欠陥信号Sa,Sb は、欠陥の種類によ
る指向性と、さまざまのサイズのために、それぞれの強
度ia,ib は大幅に変化する。
In FIG. 2, a small incident angle θ 1 and a large incident angle θ 2 are made from the back surface side of the substrate 1 to project, for example, a white light beam and a red light beam having different wavelength bands, respectively.
If there is a defect K on the surface of the substrate 1 or a defect K ′ on the back surface, these scatter scattered light. A light receiving system whose optical axis is perpendicular to the surface is provided on the surface side, and the light receiving viewing angle θ 3 of the condensing lens is narrowed so that both light fluxes transmitted through the substrate 1 are not incident, and the defect K or K ' Only the scattered light is condensed, divided by the half mirror HM, one of the transmitted light is removed by the filter F A for red color, the other reflected light is transmitted by the filter F B for red color, and each is transmitted through the CCD image sensor ( Light is received by A) and (B). Both CCD (A),
The defect signals S a and S b output by (B) have their respective intensities i a and i b significantly changed due to the directivity depending on the type of defect and various sizes.

【0007】以上の欠陥信号Sa,Sb に対して、この発
明の発明者らにより種類別の欠陥サンプルを使用して試
行実験が行われた。まず、各サンプルの欠陥信号Sa,S
b の強度比ia /ib をとると、各欠陥のサイズの相違
や、各光束の強度変動による両欠陥信号Sa とSb の強
度変化が相殺されて、指向性の特徴が抽出される。つい
で、強度比ia /ib と欠陥種別との関係を検討したと
ころ、ia /ib が一定値R1 より大きい欠陥は、ほと
んどスクラッチ傷Ks または気泡Kb であり、一定値R
1 以下のときは、ほとんど異物Ki またはスクラッチ傷
s であることが見出された。実験をさらに進めて、検
出信号Sa またはSb を画像処理して、CCDの受光し
た各素子eより欠陥Kの長さLと幅W(図2の欄外に例
示)を算出し、その比L/Wをとって欠陥種別との関係
を検討した結果、L/Wが一定値R2 より大きいとき
は、ほとんどスクラッチ傷Ks であり、一定値R2 以下
のときは、ほとんど異物Ki または気泡Kb であること
が認められた。
For the above defect signals S a and S b , the inventors of the present invention conducted a trial experiment using defect samples of each type. First, the defect signals S a and S of each sample
When the intensity ratio i a / i b of b is taken, the difference in the size of each defect and the intensity change of both defect signals S a and S b due to the intensity variation of each luminous flux are canceled out, and the directional characteristic is extracted. It Next, the relationship between the intensity ratio i a / i b and the defect type was examined. As a result, most of the defects in which i a / i b is larger than the constant value R 1 are scratches K s or bubbles K b , and the constant value R 1 is constant.
It was found that when it was 1 or less, it was mostly foreign matter K i or scratches K s . By further advancing the experiment, the detection signal S a or S b is image-processed, and the length L and width W (illustrated in the margin of FIG. 2) of the defect K are calculated from each element e received by the CCD, and the ratio thereof is calculated. As a result of examining the relationship with the defect type by taking L / W, when L / W is larger than a constant value R 2 , there are almost scratch scratches K s , and when L / W is less than the constant value R 2 , almost foreign matter K i. Alternatively, it was confirmed that the bubbles were K b .

【0008】上記の両実験の結果をまとめると、上記の
一定値R1,R2 を判定基準として、前記した各判定条件
(1) 〜(4) がえられ、これらにより、欠陥が異物Ki、ス
クラッチ傷Ks、気泡Kb のいずれであるかをほぼ確実に
判定できることが判明した。図2は各判定条件(1) 〜
(4) の実行順序を示し、まず強度比較により、ia /i
b とR1 を比較して(Ks OR Kb )であるか、もし
くは(Ki OR Ks)であるかを判定し、さらにL/
W比較により、それぞれに対するL/WとR2の比較を
行って、Ki,Ks,Kb が判定される。図3は、上記によ
り判定された各欠陥の分布領域を示すもので、ia /i
b を縦軸、L/Wを横軸とし、判定基準R1,R2 に対し
て異物Ki、スクラッチ傷Ks、気泡Kb は図示のように分
布する。
Summarizing the results of both of the above-mentioned experiments, the above-mentioned respective judgment conditions are set with the above-mentioned constant values R 1 and R 2 as judgment criteria.
(1) to (4) were obtained, and it was found that it is possible to almost certainly determine whether the defect is the foreign matter K i , the scratch K s , or the bubble K b . Figure 2 shows each judgment condition (1)
The execution order of (4) is shown. First, by strength comparison, i a / i
b and R 1 are compared to determine whether (K s OR K b ) or (K i OR K s ), and L /
By W comparison, L / W and R 2 are compared with each other, and K i , K s , and K b are determined. FIG. 3 shows the distribution area of each defect determined as described above, where i a / i
With b as the ordinate and L / W as the abscissa, the foreign matter K i , scratches K s , and bubbles K b are distributed as shown with respect to the criteria R 1 and R 2 .

【0009】さて、上記の欠陥種別判定方法には付記す
べき欠点がある。すなわち、この発明は、ガラス基板1
の表面、裏面、または内部に存在する3種の欠陥を検出
して、これらの種別を判定するものである。しかしなが
ら、基板の裏面にはTFTが形成されないので、裏面の
欠陥は検出と種別の判定は無用であり、これが裏面側で
あることを識別しない限り、基板1の管理上はむしろ有
害である。このような欠点に対して、この発明の発明者
により、表面の欠陥と裏面の欠陥とを識別する手段が考
案されて、別途、「ガラス基板の表裏欠陥識別方法」と
して特許出願される。この表裏欠陥識別方法を、上記に
より欠陥種別が判定された基板1に適用することによ
り、各欠陥(内部の気泡を除く)は表面側か裏面側かが
識別されて、欠陥検査が完了するわけである。
Now, the above-mentioned defect type judging method has a drawback to be noted. That is, the present invention relates to the glass substrate 1
Three types of defects existing on the front surface, the back surface, or inside are detected and their types are determined. However, since the TFT is not formed on the back surface of the substrate, it is unnecessary to detect the defect on the back surface and to judge the type, and it is rather harmful for the management of the substrate 1 unless it is identified that this is the back surface side. With respect to such a defect, the inventor of the present invention has devised a means for discriminating between a front surface defect and a back surface defect, and separately filed a patent as "a front and back surface defect detection method for a glass substrate". By applying this front / back defect identification method to the substrate 1 whose defect type has been determined as described above, each defect (excluding internal bubbles) is identified as the front surface side or the back surface side, and the defect inspection is completed. Is.

【0010】[0010]

【実施例】図4および図5はこの発明の一実施例を示
し、図4はこの発明の欠陥種別判定方法を適用した欠陥
検出・種別判定装置10の構成図、図5は欠陥種別判定
手順を示すフローチャートである。図4において、欠陥
検出・種別判定装置10は、検出光学系2、XYステー
ジ3、データ処理部4とにより構成される。基板1はX
Yステージ3に載置され、XY移動制御回路3a により
一端から他端までがX方向にステップ移動し、ついでY
方向に順次に移動してX方向のステップ移動が繰り返さ
れて全面が検査される。検出光学系2は投光系21と受光
系22よりなり、投光系21は基板1の裏面側に設けられ、
ハロゲンランプよりなる4個の光源211,212,212,212 を
有し、光源211の白色光束は集束レンズ213 により集束
されて、基板1の検査位置に対して入射角θ1 を約20
°として投射され、また他の3個の各光源212 の白色光
束は、それぞれの集束レンズ214 により集束され、フィ
ルタ215 により赤色を選択して、検査位置に対して入射
角θ2 を約60°として3方向より投射される。基板1
のステップ移動により検査位置に来た表面欠陥Kまたは
裏面欠陥K’は、光束の散乱光をそれぞれ散乱する。
4 and 5 show an embodiment of the present invention, FIG. 4 is a block diagram of a defect detection / type determination device 10 to which the defect type determination method of the present invention is applied, and FIG. 5 is a defect type determination procedure. It is a flowchart showing. In FIG. 4, the defect detection / type determination device 10 includes a detection optical system 2, an XY stage 3, and a data processing unit 4. Substrate 1 is X
It is placed on the Y stage 3, and one end to the other end is step-moved in the X direction by the XY movement control circuit 3a.
Direction is sequentially moved, and step movement in the X direction is repeated to inspect the entire surface. The detection optical system 2 includes a light projecting system 21 and a light receiving system 22, and the light projecting system 21 is provided on the back surface side of the substrate 1.
It has four light sources 211, 212, 212, 212 composed of halogen lamps, and the white light flux of the light source 211 is focused by the focusing lens 213 so that the incident angle θ 1 with respect to the inspection position of the substrate 1 is about 20.
The white light flux of each of the other three light sources 212 is focused by each focusing lens 214, red is selected by the filter 215, and the incident angle θ 2 is about 60 ° with respect to the inspection position. Is projected from three directions. Substrate 1
The surface defect K or the back surface defect K ′ that has come to the inspection position due to the step movement of 1 scatters the scattered light of the light flux.

【0011】受光系22は基板1の表面側に、光軸を基板
1に対して垂直として設けられ、その集光レンズ221
は、基板1を透過した各光束が入射しないように、受光
視角θ3 を約15°として欠陥K,K’の散乱光のみを
集光する。この場合、光束は4方向より投射されるの
で、各欠陥の散乱光には垂直上方を向く成分があり、こ
れが集光レンズ221 にかならず入射する。集光された散
乱光はハーフミラー222 により分割され、一方はフィル
タ223 により赤色光が除去されてCCD(A)224に受
光され、他方はフィルタ225 により赤色光が透過してC
CD(B)226 に受光され、それぞれより欠陥信号Sa,
b が出力される。データ処理部4は、画像メモリ41、
マイクロプロセッサ(MPU)42、プリンタ43などより
なる。画像メモリ41には、基板1のステップ移動により
両CCD(A),(B)が出力する欠陥信号Sa,Sb が、
検査位置ごとに更新記憶される。MPU42は、強度比較
プログラムPG1、画像処理プログラムPG2、L/W
比較プログラムPG3、サイズ算出プログラムPG4、
制御信号発生プログラムPG5、データ編集プログラム
PG6を有し、PG1とPG3には、予め試行実験など
によりえられた判定基準R1,R2 がそれぞれ設定され
る。PG5は制御信号を発生し、これをXY移動制御回
路3a に与えて上記した基板1のステップ移動を行う。
The light receiving system 22 is provided on the front surface side of the substrate 1 with its optical axis perpendicular to the substrate 1, and its condensing lens 221.
Collects only the scattered light of the defects K and K ′ with the light-receiving viewing angle θ 3 set to about 15 ° so that each light beam transmitted through the substrate 1 does not enter. In this case, since the light beam is projected from four directions, the scattered light of each defect has a component directed vertically upward, and this component always enters the condenser lens 221. The scattered light thus collected is split by the half mirror 222, one of which removes the red light by the filter 223 and is received by the CCD (A) 224, and the other of which is transmitted by the filter 225 to the red light and C
The CD (B) 226 receives the light, and the defect signals S a ,
S b is output. The data processing unit 4 includes an image memory 41,
It includes a microprocessor (MPU) 42, a printer 43, and the like. In the image memory 41, the defect signals S a and S b output by both CCDs (A) and (B) by the step movement of the substrate 1,
It is updated and stored for each inspection position. The MPU 42 includes an intensity comparison program PG1, an image processing program PG2, and L / W.
Comparison program PG3, size calculation program PG4,
It has a control signal generation program PG5 and a data editing program PG6, and the judgment criteria R 1 and R 2 obtained in advance by trial experiments are set in PG1 and PG3, respectively. The PG 5 generates a control signal and supplies it to the XY movement control circuit 3a to perform the step movement of the substrate 1 described above.

【0012】以下、図4に図5を併用してデータ処理部
4における欠陥種別判定手順を説明する。画像メモリ41
に更新記憶されている両欠陥信号Sa,Sb は、MPU42
に読出されて、PG1の処理により強度比ia /ib
算出され(フローチャートのステップ)、これに設定
されている判定基準R1 に比較される。比較結果がi
a /ib >R1 ;(YES)のときは(Ks OR Kb)
と判定され、その欠陥信号Sa またはSb がPG2によ
り処理されて、欠陥の長さと幅の比L/Wが算出され
、これがPG3により判定基準R2 と比較される。
L/W≦R2 ;(NO)のときはKb 、L/W>R2
(YES)のときはKs と判定され、それぞれのサイズ
がPG4により算出され、これにXY移動制御回路3
a より与えられる欠陥のXY座標値が付加され、これ
らと欠陥種別とがPG6により編集されて、プリンタ
43に出力されてプリントされる。ステップにおいて、
a /ib ≦R1 ;(NO)のときは(Ki OR Ks)
と判定され、上記と同様に、L/Wの算出と、L/W
とR2 の比較とがなされ、L/W>R2 ;(YES)
のときはKs 、L/W≦R2 ;(NO)のときはKi
判定され、上記と同様に、それぞれのサイズの算出、
XY座標値の付加、データ編集が順次になされてプ
リンタ43にプリントされる。
The defect type determination procedure in the data processing unit 4 will be described below with reference to FIG. 4 and FIG. Image memory 41
Both defect signals S a and S b updated and stored in
Is read out and the intensity ratio i a / i b is calculated by the processing of PG1 (step in the flowchart) and compared with the determination standard R 1 set therein. The comparison result is i
a / i b > R 1 ; (YES) (K s OR K b ).
Then, the defect signal S a or S b is processed by the PG2 to calculate the defect length-width ratio L / W, which is compared with the judgment criterion R 2 by the PG3.
L / W≤R 2 ; Kb when (NO), L / W> R 2 ;
When (YES), it is determined to be K s , each size is calculated by PG4, and the XY movement control circuit 3
The XY coordinate values of the defect given by a are added, and these and the defect type are edited by PG6,
It is output to 43 and printed. In the step
i a / i b ≤R 1 ; when (NO), (K i OR K s ).
It is determined that L / W is calculated and L / W is calculated in the same manner as above.
And R 2 are compared, and L / W> R 2 ; (YES)
Is determined to be K s , L / W ≦ R 2 ; (NO) is determined to be K i .
The XY coordinate values are added and the data is edited in sequence and printed on the printer 43.

【0013】[0013]

【発明の効果】以上の説明のとおり、この発明の欠陥種
別判定方法は、ガラス基板に対して異なる入射角で、異
なる2波長帯域の光束を投射し、各欠陥の散乱光を波長
帯域別にCCDイメージセンサにより受光し、それぞれ
の出力する欠陥信号Sa,Sb の強度比ia /ib と、各
欠陥の長さLと幅Wの比L/Wとを算出し、両比を一定
値R1,R2 にそれぞれ比較する試行実験の結果、R1,R
2 を判定基準とする判定条件がえられ、これに基づい
て、検出した3種の欠陥の種別が、異物、スクラッチ
傷、 気泡のいずれであるかを、ほぼ確実に判定して区分
できるもので、液晶用のガラス基板の製造・管理に寄与
する効果には、優れたものがある。
As described above, according to the defect type determining method of the present invention, the light beams of two different wavelength bands are projected to the glass substrate at different incident angles, and the scattered light of each defect is CCD by wavelength band. The intensity ratio i a / i b of the defect signals S a and S b received by the image sensor and output from each of them and the ratio L / W of the length L and the width W of each defect are calculated, and both ratios are constant. The results of trial experiments comparing the values R 1 and R 2 respectively, R 1 and R
Based on this, a judgment condition based on 2 is obtained, and based on this, it is possible to almost certainly judge and distinguish whether the three types of defects detected are foreign matter, scratches, or bubbles. The effects that contribute to the production and management of glass substrates for liquid crystals are excellent.

【図面の簡単な説明】[Brief description of drawings]

【図1】 図1は、3種の欠陥の形状などの説明図であ
って、(a)は、その欠陥形状の説明図、(b)は、欠陥の指
向特性の説明図である。
FIG. 1 is an explanatory diagram of shapes of three types of defects, (a) is an explanatory diagram of the defect shape, and (b) is an explanatory diagram of defect directional characteristics.

【図2】 図2は、この発明の欠陥種別判定方法の原理
の説明図である。
FIG. 2 is an explanatory diagram of the principle of the defect type determination method of the present invention.

【図3】 図3は、欠陥種別判定方法により判定された
3種類の欠陥の分布領域を示す分布図である。
FIG. 3 is a distribution diagram showing distribution areas of three types of defects determined by a defect type determination method.

【図4】 図4は、この発明を適用した欠陥検査装置の
一実施例の構成図である。
FIG. 4 is a configuration diagram of an embodiment of a defect inspection apparatus to which the present invention is applied.

【図5】 図5は、図4に対する欠陥種別判定手順を示
すフローチャートである。
FIG. 5 is a flowchart showing a defect type determination procedure for FIG.

【符号の説明】[Explanation of symbols]

1…液晶パネル用のガラス基板、2…検出光学系、21…
投光系、211,212 …ハロゲンランプの光源、213,214 …
集束レンズ、215 …赤色フィルタ、22…受光系、221 …
集光レンズ、222 …ハーフミラー、223、225 …フィル
タ、224,226 …CCDイメージセンサ、3…XYステー
ジ、3a …XY移動制御回路、4…データ処理部、41…
画像メモリ、42…MPU、43…プリンタ、10…この発
明による欠陥検査装置、K…表面の欠陥、K’…裏面の
欠陥、Ki …異物、Ks …スクラッチ傷、Kb …気泡、
θ12 …入射角、θ3 …受光視角、Sa,Sb …欠陥信
号、ia,ib …欠陥信号の強度、L…欠陥の長さ、W…
欠陥の幅、R1,R2 …判定基準、PG1〜PG6…MP
Uのプログラム、〜…フローチャートのステップ番
号。
1 ... Glass substrate for liquid crystal panel, 2 ... Detection optical system, 21 ...
Projection system, 211,212 ... Light source of halogen lamp, 213,214 ...
Focusing lens, 215 ... Red filter, 22 ... Light receiving system, 221 ...
Condensing lens, 222 ... Half mirror, 223, 225 ... Filter, 224, 226 ... CCD image sensor, 3 ... XY stage, 3a ... XY movement control circuit, 4 ... Data processing section, 41 ...
Image memory, 42 ... MPU, 43 ... Printer, 10 ... Defect inspection apparatus according to the present invention, K ... Front surface defect, K ′ ... Back surface defect, K i ... Foreign matter, K s ... Scratch scratch, K b ... Bubble,
θ 1 , θ 2 ... Incident angle, θ 3 ... Receiving viewing angle, S a , S b ... Defect signal, i a , i b ... Defect signal intensity, L ... Defect length, W ...
Defect width, R 1 , R 2 ... Judgment standard, PG1 to PG6 ... MP
U's program, ~ ... Step number in the flowchart.

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】液晶パネル用のガラス基板の、TFTを形
成する表面またはその裏面に存在する異物Ki とスクラ
ッチ傷Ks 、および該ガラス基板の内部に存在する気泡
bの3種の欠陥の種別判定において、該ガラス基板の
裏面側に設けた投光系により、波長帯域の異なる少なく
とも2本の光束を、該ガラス基板の検査位置に対して小
さい入射角と大きい入射角でそれぞれ投射し、該ガラス
基板の表面側に、光軸が該ガラス基板に垂直な受光系を
設け、該受光系が有する前記各光束の入射しない狭い受
光視角の集光レンズにより、前記各欠陥による前記各光
束の散乱光を集光し、該集光された散乱光を前記異なる
波長帯域に分割して、それぞれのCCDイメージセンサ
により受光し、該両CCDイメージセンサの出力する欠
陥信号Sa,Sb の強度ia,ib の比ia /ib と、該欠
陥信号Sa またはSb を画像処理して求めた前記各欠陥
の長さLと幅Wの比L/Wを算出し、該両比ia /i
b ,L/Wに対する判定基準をそれぞれR1,R2 とし
て、下記の各判定条件: (ia /ib >R1 ,L/W>R2 ):Ks ………条件(1) (ia /ib >R1 ,L/W≦R2 ):Kb ………条件(2) (ia /ib ≦R1 ,L/W>R2 ):Ks ………条件(3) (ia /ib ≦R1 ,L/W≦R2 ):Ki ………条件(4) により、前記異物Ki 、スクラッチ傷Ks 、気泡Kb
それぞれ判定することを特徴とする、ガラス基板の欠陥
種別判定方法。
1. A glass substrate for a liquid crystal panel, which has three types of defects: a foreign substance K i and a scratch K s existing on the front surface or the back surface of a TFT, and a bubble K b existing inside the glass substrate. In the type determination, the light projecting system provided on the back surface side of the glass substrate projects at least two light beams having different wavelength bands at a small incident angle and a large incident angle with respect to the inspection position of the glass substrate. , A light receiving system whose optical axis is perpendicular to the glass substrate is provided on the front surface side of the glass substrate, and the light receiving system has a narrow converging lens with a narrow light receiving visual angle from which each light beam is caused by each defect. Of the defect signals S a and S b output by both CCD image sensors. strength The ratio i a / i b of the degrees i a and i b and the ratio L / W of the length L and the width W of each defect obtained by image-processing the defect signal S a or S b are calculated. Both ratios i a / i
b, and criteria for L / W as R 1, R 2, respectively, each determined under the following conditions: (i a / i b> R 1, L / W> R 2): K s ......... condition (1) (i a / i b> R 1, L / W ≦ R 2): K b ......... condition (2) (i a / i b ≦ R 1, L / W> R 2): K s ......... Condition (3) (i a / i b ≤R 1 , L / W ≤R 2 ): K i ............ The condition K4 determines the foreign matter K i , scratch scratch K s , and bubble K b , respectively. A method of determining a defect type of a glass substrate, comprising:
JP8087049A 1996-03-18 1996-03-18 Glass substrate defect type determination method Pending JPH09257642A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP8087049A JPH09257642A (en) 1996-03-18 1996-03-18 Glass substrate defect type determination method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP8087049A JPH09257642A (en) 1996-03-18 1996-03-18 Glass substrate defect type determination method

Publications (1)

Publication Number Publication Date
JPH09257642A true JPH09257642A (en) 1997-10-03

Family

ID=13904096

Family Applications (1)

Application Number Title Priority Date Filing Date
JP8087049A Pending JPH09257642A (en) 1996-03-18 1996-03-18 Glass substrate defect type determination method

Country Status (1)

Country Link
JP (1) JPH09257642A (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001235427A (en) * 2000-02-24 2001-08-31 Topcon Corp Surface inspection equipment
JP2005201887A (en) * 2003-12-16 2005-07-28 Hitachi High-Tech Electronics Engineering Co Ltd Glass substrate inspection method and inspection apparatus, and display panel manufacturing method
JP2007024733A (en) * 2005-07-19 2007-02-01 Hitachi High-Technologies Corp Substrate inspection apparatus and substrate inspection method
JP2007199066A (en) * 2006-01-26 2007-08-09 Orbotech Ltd System and method for inspecting patterned devices with fine conductors
JP2007212479A (en) * 2007-05-07 2007-08-23 Hitachi Ltd Defect inspection apparatus and method
JP2008076071A (en) * 2006-09-19 2008-04-03 Hitachi High-Technologies Corp Substrate inspection apparatus and substrate inspection method
JP2009516852A (en) * 2005-11-21 2009-04-23 コーニング インコーポレイテッド Inclined transmitted illumination inspection system and method for inspecting glass sheet
JP2011017708A (en) * 2010-08-04 2011-01-27 Hitachi High-Technologies Corp Substrate-inspecting device and substrate inspection method
JP2012096920A (en) * 2010-11-05 2012-05-24 Hitachi High-Technologies Corp Glass substrate defect inspection device and glass substrate defect inspection method and glass substrate defect inspection system
KR101496993B1 (en) * 2013-09-02 2015-03-02 (주) 인텍플러스 inspection method for display panel
CN114608818A (en) * 2022-03-01 2022-06-10 深圳市华星光电半导体显示技术有限公司 Folding screen crease detection device and method thereof

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001235427A (en) * 2000-02-24 2001-08-31 Topcon Corp Surface inspection equipment
JP2005201887A (en) * 2003-12-16 2005-07-28 Hitachi High-Tech Electronics Engineering Co Ltd Glass substrate inspection method and inspection apparatus, and display panel manufacturing method
JP2007024733A (en) * 2005-07-19 2007-02-01 Hitachi High-Technologies Corp Substrate inspection apparatus and substrate inspection method
JP2009516852A (en) * 2005-11-21 2009-04-23 コーニング インコーポレイテッド Inclined transmitted illumination inspection system and method for inspecting glass sheet
JP2007199066A (en) * 2006-01-26 2007-08-09 Orbotech Ltd System and method for inspecting patterned devices with fine conductors
JP2008076071A (en) * 2006-09-19 2008-04-03 Hitachi High-Technologies Corp Substrate inspection apparatus and substrate inspection method
JP2007212479A (en) * 2007-05-07 2007-08-23 Hitachi Ltd Defect inspection apparatus and method
JP2011017708A (en) * 2010-08-04 2011-01-27 Hitachi High-Technologies Corp Substrate-inspecting device and substrate inspection method
JP2012096920A (en) * 2010-11-05 2012-05-24 Hitachi High-Technologies Corp Glass substrate defect inspection device and glass substrate defect inspection method and glass substrate defect inspection system
KR101496993B1 (en) * 2013-09-02 2015-03-02 (주) 인텍플러스 inspection method for display panel
CN114608818A (en) * 2022-03-01 2022-06-10 深圳市华星光电半导体显示技术有限公司 Folding screen crease detection device and method thereof
CN114608818B (en) * 2022-03-01 2023-12-01 深圳市华星光电半导体显示技术有限公司 Folding screen crease detection device and method thereof

Similar Documents

Publication Publication Date Title
KR102038478B1 (en) Wafer inspection method and wafer inspection apparatus
TWI713638B (en) Method for detecting defects and associated device
JP2002529698A (en) Glass inspection equipment
JP3938227B2 (en) Foreign object inspection method and apparatus
JPH09257642A (en) Glass substrate defect type determination method
JP2009042202A (en) Wafer inspection apparatus and wafer inspection method
JP4158227B2 (en) Inspection method and inspection apparatus for minute unevenness
US8184282B2 (en) Method and system for defect detection using transmissive bright field illumination and transmissive dark field illumination
JP3480176B2 (en) Glass substrate front / back defect identification method
JP4151306B2 (en) Inspection method of inspection object
JPH08271438A (en) Inspection equipment
JP2006017685A (en) Surface defect inspection device
JP4408902B2 (en) Foreign object inspection method and apparatus
JPS59135353A (en) Surface flaw detecting apparatus
US5745239A (en) Multiple focal plane image comparison for defect detection and classification
JP3100449B2 (en) Surface condition inspection device
JP2010190740A (en) Substrate inspection device, method, and program
JPH0311403B2 (en)
JPH10267858A (en) Glass substrate defect judgment method
JP3149336B2 (en) Optical member inspection device
JPH07159333A (en) Appearance inspection device and appearance inspection method
JPH11281588A (en) Surface inspecting apparatus
JPH10206341A (en) Foreign object detection device in nonmetallic inclusion measurement
JPH10267857A (en) Method for determining the continuity of scratches
JP3197047B2 (en) Defect inspection equipment