JP2003329609A - Method of deleting pattern and method of inspecting defect - Google Patents
Method of deleting pattern and method of inspecting defectInfo
- Publication number
- JP2003329609A JP2003329609A JP2002134189A JP2002134189A JP2003329609A JP 2003329609 A JP2003329609 A JP 2003329609A JP 2002134189 A JP2002134189 A JP 2002134189A JP 2002134189 A JP2002134189 A JP 2002134189A JP 2003329609 A JP2003329609 A JP 2003329609A
- Authority
- JP
- Japan
- Prior art keywords
- pattern
- image
- pitch
- natural number
- inspected
- 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.)
- Granted
Links
Landscapes
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Liquid Crystal (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
- Solid State Image Pick-Up Elements (AREA)
Abstract
Description
【０００１】[0001]
【発明の属する技術分野】本発明は主に液晶パネル、プ
ラズマディスプレイ、半導体ウェハ等の電子機器デバイ
スを撮像し、その撮像画像中に含まれる繰り返しパター
ンを消去し、繰り返しパターン中の欠陥判定を行う、撮
像画像中の繰り返しパターン消去方法および欠陥検査方
法に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention mainly takes an image of an electronic device device such as a liquid crystal panel, a plasma display, a semiconductor wafer, etc., erases a repetitive pattern included in the picked-up image, and judges a defect in the repetitive pattern. The present invention relates to a repeated pattern erasing method and a defect inspection method in a captured image.
【０００２】[0002]
【従来の技術】液晶パネル、プラズマディスプレイ、半
導体ウェハ等の電子機器デバイスの繰り返しパターンの
欠陥検査を行うため、部分的に同じパターンが周期的に
繰り返し全体に形成されている被検査物を撮像する。そ
の撮像画像の繰り返しパターンの周期性を満たさない部
分を検出するため、被検査物を撮像した画素のうち、注
目画素とパターンの繰り返し周期分離れた位置に存在す
る画素を比較していた。具体的には、注目画素とパター
ンのピッチの整数倍の周期分離れた位置に存在する画素
の濃度の差分を取り、その差分を取った濃度値が第１の
所定の値以上または第２の所定の値以下のところを、夫
々周期性を満たさない部分として欠陥としていた。2. Description of the Related Art In order to perform a defect inspection of a repetitive pattern of an electronic device such as a liquid crystal panel, a plasma display, a semiconductor wafer, etc., an image of an object to be inspected in which a part of the same pattern is periodically repeated is formed. . In order to detect a portion of the captured image that does not satisfy the periodicity of the repetitive pattern, the pixel of interest is compared with the pixel existing at a position separated from the pattern by the repetitive cycle. Specifically, the difference between the densities of the pixel of interest and a pixel existing at a position separated by a period that is an integral multiple of the pitch of the pattern is calculated, and the density value obtained by the difference is equal to or greater than a first predetermined value or a second density value. The areas below the predetermined value were regarded as defects as the areas that did not satisfy the periodicity.
【０００３】図４に従来のパターン消去方法を示す。図
４（ａ）において、１は被検査物を撮像した画像であ
る。画像１には、繰り返しパターン２と、黒欠陥３ａ、
白欠陥３ｂが撮像されている。４は画像１のチェックラ
インであり、チェックライン４の線上の画像データの濃
淡を示すのが濃度プロファイルｆ１である。ここで繰り
返しパターンのピッチｐの間隔に濃度プロファイルｆ１
をｃ１、ｃ２、ｃ３、ｃ４に分割する。FIG. 4 shows a conventional pattern erasing method. In FIG. 4A, reference numeral 1 is an image obtained by imaging the inspection object. In the image 1, the repeating pattern 2 and the black defect 3a,
The white defect 3b is imaged. Reference numeral 4 denotes a check line of the image 1, and the density profile f1 indicates the density of the image data on the check line 4. Here, the density profile f1 is arranged at intervals of the pitch p of the repeating pattern.
Is divided into c1, c2, c3 and c4.
【０００４】そして、ｃ２の濃淡値からｃ１濃淡値を引
き、ｃ３の濃淡値からｃ２濃淡値を引き、ｃ４の濃淡値
からｃ３濃淡値を引く。こうして濃度プロファイルｆ１
についてピッチｐ離れた濃淡値の差分をとる。つぎに順
次、画像チェックライン４を上下に移動し、濃度プロフ
ァイルを分割し差分をとり、画像１全面に対し上記の方
法で濃度値の差分をとる。Then, the gray value of c1 is subtracted from the gray value of c2, the gray value of c2 is subtracted from the gray value of c3, and the gray value of c3 is subtracted from the gray value of c4. Thus, the density profile f1
Is calculated by taking a difference between the grayscale values separated by the pitch p. Next, the image check line 4 is sequentially moved up and down, the density profile is divided and the difference is taken, and the difference of the density value is taken for the entire surface of the image 1 by the above method.
【０００５】こうして画像１全面の差分をとり、図４
（ｂ）に示す様に、被検査物を撮像した画像１のパター
ンを消去した画像が６である。次に、パターン消去画像
６に順次チェックラインを設け、黒欠陥７ａと白欠陥７
ｂを判定する。例えば、パターン消去画像６上にチェッ
クライン８を設けると、チェックライン８の線上のデー
タの濃淡を示す差分濃度プロファイルはｆ２の様にな
る。Thus, the difference of the entire image 1 is obtained,
As shown in (b), the image 6 in which the pattern of the image 1 obtained by imaging the inspection object is deleted is 6. Next, check lines are sequentially provided on the pattern erased image 6, and the black defect 7a and the white defect 7 are
Determine b. For example, when the check line 8 is provided on the pattern erased image 6, the difference density profile showing the light and shade of the data on the line of the check line 8 becomes f2.
【０００６】ここで、１０は濃度値が第１の所定の値で
ある白欠陥閾値、１１は濃度値が第２の所定の値である
黒欠陥閾値である。差分濃度プロファイルｆ２の１２ａ
は濃度値が低く黒欠陥閾値１１を下回っているため、１
２ａは黒欠陥となる。差分濃度プロファイルｆ２の１２
ｂは濃度値が高く白欠陥閾値１０を上回っているため、
１２は白欠陥となる。同様に、パターン消去画像６に順
次チェックラインを設け、パターン消去画像６全面の欠
陥検査を行う。Here, 10 is a white defect threshold value whose density value is a first predetermined value, and 11 is a black defect threshold value whose density value is a second predetermined value. 12a of difference density profile f2
Has a low density value and is below the black defect threshold value 11, so 1
2a becomes a black defect. 12 of the difference density profile f2
Since b has a high density value and exceeds the white defect threshold value 10,
12 is a white defect. Similarly, a check line is sequentially provided on the pattern erased image 6, and a defect inspection of the entire pattern erased image 6 is performed.
【０００７】[0007]
【発明が解決しようとする課題】しかしながら、上記従
来の方法では、撮像手段がＣＣＤ等の固体撮像素子の場
合、被検査物を撮像した画像には量子化誤差が発生す
る。特に固体撮像素子に結像された被検査物のパターン
の線幅が１画素のサイズに近い場合、その影響は顕著で
あり、結像したパターンと画素の位置関係により、撮像
画像上でパターンの幅と濃度は大きく変化する。However, in the above-mentioned conventional method, when the image pickup means is a solid-state image pickup device such as a CCD, a quantization error occurs in the image picked up from the inspection object. In particular, when the line width of the pattern of the inspection object imaged on the solid-state image sensor is close to the size of one pixel, the effect is remarkable, and the positional relationship between the imaged pattern and the pixel causes the pattern of the pattern to appear on the captured image. The width and density vary greatly.
【０００８】図５にその理由を示す。量子化誤差がない
理想的な状態では、被検査物を撮像した画像２０の理想
濃度プロファイルはｆ３の様にどのパターン２に対して
も一様な濃度となる。しかし、実際はＣＣＤ（固体撮像
素子）２２で被検査物を撮像するためＣＣＤ２２の１画
素単位で量子化される。この撮像画像２０のＣＣＤ２２
でのパターン２のＣＣＤ濃度プロファイルｆ４は、パタ
ーン２とＣＣＤ２２の画素の位置関係で濃度が変化する
ため、理想の濃度プロファイルｆ３とは異なった濃度値
を示す。The reason for this is shown in FIG. In an ideal state where there is no quantization error, the ideal density profile of the image 20 of the object to be inspected is uniform for any pattern 2 like f3. However, in practice, the CCD (solid-state image sensor) 22 images the object to be inspected, so that the CCD 22 is quantized in pixel units. CCD 22 of this captured image 20
The CCD density profile f4 of the pattern 2 in (1) shows a density value different from the ideal density profile f3 because the density changes depending on the positional relationship between the pattern 2 and the pixels of the CCD 22.
【０００９】よって、上記従来の技術の方法で、ＣＣＤ
濃度プロファイルｆ４について、パターンの１ピッチ分
離れた画素の差分をとると、差分濃度プロファイルはｆ
５に示す様な複雑な線、つまりノイズの多い差分濃度プ
ロファイルｆ５になってしまう。このため、差分濃度プ
ロファイルｆ５が、白欠陥閾値１０を上回る部分２５は
白欠陥、黒欠陥閾値１１を下回る部分２６は黒欠陥、と
なる。つまり、被検査物の撮像画像上に欠陥が無くと
も、欠陥を検出してしまう課題があった。Therefore, according to the above conventional method, the CCD
For the density profile f4, the difference density profile is f
5, a complicated line as shown in FIG. 5, that is, a noisy difference density profile f5 is obtained. Therefore, the portion 25 in which the difference density profile f5 exceeds the white defect threshold value 10 is a white defect, and the portion 26 in which the difference density profile f5 is less than the black defect threshold value 11 is a black defect. That is, even if there is no defect on the captured image of the inspection object, there is a problem that the defect is detected.
【００１０】[0010]
【課題を解決するための手段】上記課題を解決するため
に本発明のパターン消去方法は、固体撮像素子の画素の
ピッチをｃとし、パターンのピッチをｐとし、レンズの
倍率をｍとし、自然数をｎとし、自然数をｊとし、ｋ＝
ｎ×ｍ×ｐ／ｃの式でピッチ係数ｋを定義し、ピッチ係
数ｋの小数点以下を四捨五入した数をｉとし、ピッチ係
数ｋが０．９×ｉ≦ｋ≦１．１×ｉの範囲になるように
前記レンズの倍率ｍと自然数ｎを設定し、ｎ×ｊ周期離
れている撮像画像のパターンの差分をとり、撮像画像全
体のパターンを消去する事を特徴とする。In order to solve the above problems, the pattern erasing method of the present invention is such that the pixel pitch of the solid-state image pickup device is c, the pattern pitch is p, the lens magnification is m, and a natural number. Is n, a natural number is j, and k =
The pitch coefficient k is defined by the formula of n × m × p / c, the number rounded off the decimal point of the pitch coefficient k is i, and the pitch coefficient k is in the range of 0.9 × i ≦ k ≦ 1.1 × i. The magnification m and the natural number n of the lens are set so that the difference between the patterns of the captured images separated by n × j cycles is taken, and the pattern of the entire captured image is erased.
【００１１】また本発明の欠陥検査方法は、上記パター
ン消去方法を用い、パターンを消去した撮像画像の濃度
値が第１の所定の値以上または第２の所定の値以下の部
分を夫々欠陥とする事を特徴とする。According to the defect inspection method of the present invention, the above-mentioned pattern erasing method is used, and a portion where the density value of the captured image from which the pattern is erased is equal to or higher than the first predetermined value or equal to or lower than the second predetermined value is regarded as a defect. It is characterized by doing.
【００１２】さらに、上記欠陥検査方法を、液晶パネ
ル、プラズマディスプレイパネル、あるいは半導体ウェ
ハの欠陥検査方法に用いてもよい。Further, the above defect inspection method may be used for a defect inspection method for a liquid crystal panel, a plasma display panel, or a semiconductor wafer.
【００１３】これにより、被検査物を固体撮像素子にて
撮像した時でも量子化誤差の少ない撮像画像を得ること
が可能になり、適切に撮像画像のパターンを消去し、正
確に欠陥を検査することが可能になる。Thus, it becomes possible to obtain a picked-up image with a small quantization error even when the object to be inspected is picked up by the solid-state image pickup element, the pattern of the picked-up image is appropriately erased, and the defect is accurately inspected. It will be possible.
【００１４】[0014]
【発明の実施の形態】図１〜図３に本発明の実施の形態
を示す。図１は本実施の形態のパターン消去方法を実現
する撮像装置の概念図である。図１において、３１は被
検査物であり、具体的には表面にストライプ状のパター
ンを有する液晶用のガラス基板である。1 to 3 show an embodiment of the present invention. FIG. 1 is a conceptual diagram of an image pickup apparatus that realizes the pattern erasing method of the present embodiment. In FIG. 1, reference numeral 31 denotes an object to be inspected, specifically, a glass substrate for liquid crystal having a striped pattern on the surface.
【００１５】この被検査物３１の像はレンズ３２を通し
てカメラ３３中のＣＣＤ（固体撮像素子）３４上に結像
される。このＣＣＤ３４により撮像した被検査物３１の
撮像画像は、制御装置３５に撮像データとして記憶され
る。また制御装置３５によりレンズ３２の倍率は変更可
能になっている。An image of the inspection object 31 is formed on a CCD (solid-state image pickup device) 34 in a camera 33 through a lens 32. The captured image of the inspection object 31 captured by the CCD 34 is stored in the control device 35 as captured image data. Further, the magnification of the lens 32 can be changed by the control device 35.
【００１６】図２に撮像画像をＣＣＤの画素により量子
化する概念図を示す。４１は被検査物を撮像した画像で
あり、４２はストライプ状のパターンである。従来の技
術と同様に画像４１にチェックライン４３をもうけ、Ｃ
ＣＤの画素４４によりチェックライン４３線上の画像の
濃淡を検出する。このチェックライン４３線上の画像の
濃淡を示すのが濃度プロファイルｆ６である。FIG. 2 shows a conceptual diagram of quantizing the picked-up image by the pixels of the CCD. Reference numeral 41 is an image obtained by picking up an object to be inspected, and 42 is a striped pattern. As in the conventional technique, a check line 43 is added to the image 41, and C
The density of the image on the check line 43 is detected by the pixel 44 of the CD. The density profile f6 indicates the density of the image on the check line 43.
【００１７】つぎに量子化誤差を防止し、パターンを検
出する方法を示す。被検査物のパターンのピッチをＰと
し、固体撮像素子（ＣＣＤ）の画素のピッチをｃとし、
レンズの倍率をｍとし、ピッチ係数をｋｆとし、式
（１）を定義する。Next, a method of preventing a quantization error and detecting a pattern will be described. The pattern pitch of the object to be inspected is P, the pixel pitch of the solid-state image sensor (CCD) is c,
Equation (1) is defined with the lens magnification being m and the pitch coefficient being kf.
【００１８】[0018]
【数１】 [Equation 1]
【００１９】この式（１）でピッチ係数ｋｆが整数とな
る様に、レンズの倍率ｍを設定する。この整数は、小さ
い方が撮像画像のパターンを消去しやすい。一例とし
て、被検査物のパターンのピッチＰが２８μｍ、固体撮
像素子（ＣＣＤ）の画素のピッチｃが１２μｍのとき、
レンズ倍率ｍを３とする。このとき、ピッチ係数ｋｆは
７となり、図２に示すように、固体撮像素子（ＣＣＤ）
の画素のピッチｃの整数倍（７倍）が、撮像画像４１上
のパターン４２のピッチ間距離となる。つまり、撮像画
像４１上のどの位置のパターン４２でも、固体撮像素子
（ＣＣＤ）４４のパターンを量子化する画素との位置関
係が一定になる。これにより撮像画像のパターンを固体
撮像素子（ＣＣＤ）にて量子化した時の誤差は無くな
り、パターンに対する濃度プロファイルｆ６の濃淡は常
に一定となる。In the equation (1), the magnification m of the lens is set so that the pitch coefficient kf is an integer. The smaller this integer is, the easier the pattern of the captured image is erased. As an example, when the pattern pitch P of the inspection object is 28 μm and the pixel pitch c of the solid-state image sensor (CCD) is 12 μm,
The lens magnification m is 3. At this time, the pitch coefficient kf becomes 7, and as shown in FIG. 2, the solid-state image sensor (CCD)
An integer multiple (7 times) of the pitch c of the pixels is the inter-pitch distance of the pattern 42 on the captured image 41. That is, the positional relationship between the pattern 42 at any position on the captured image 41 and the pixel that quantizes the pattern of the solid-state image sensor (CCD) 44 becomes constant. This eliminates the error when the pattern of the captured image is quantized by the solid-state image sensor (CCD), and the density of the density profile f6 with respect to the pattern is always constant.
【００２０】この状態で、従来の技術と同様に濃度プロ
ファイルｆ６に対しパターンのピッチｐ離れた濃淡値の
差分をとり、被検査物を撮像した画像のパターンを消去
する。In this state, as in the prior art, the difference between the density values at the pattern pitch p with respect to the density profile f6 is calculated, and the pattern of the image of the object to be inspected is erased.
【００２１】しかし、レンズの倍率ｍの範囲は制約があ
り、被検査物のパターンのピッチＰと、固体撮像素子
（ＣＣＤ）の画素のピッチｃの数値によっては、上記実
施の形態１の式（１）で示したピッチ係数ｋｆは、整数
にはならない場合がある。一例として、被検査物のパタ
ーンのピッチＰが２２μｍ、固体撮像素子（ＣＣＤ）の
画素のピッチｃが１２μｍのとき、レンズ倍率ｍを３と
する。このとき、ピッチ係数ｋｆは５．５となり、整数
ではなくなる。このため図３に示す様に、隣接するパタ
ーンの差分をとっても、正確にパターンを消去できな
い。However, the range of the magnification m of the lens is limited, and depending on the numerical value of the pitch P of the pattern of the object to be inspected and the pixel pitch c of the solid-state image pickup device (CCD), the equation (1) of the first embodiment is used. The pitch coefficient kf shown in 1) may not be an integer. As an example, when the pattern pitch P of the inspection object is 22 μm and the pixel pitch c of the solid-state image sensor (CCD) is 12 μm, the lens magnification m is 3. At this time, the pitch coefficient kf becomes 5.5 and is not an integer. Therefore, as shown in FIG. 3, even if the difference between adjacent patterns is taken, the patterns cannot be erased accurately.
【００２２】ここでｎを自然数とし、式（１）を変更
し、ピッチ係数ｋをつぎの式（２）で定義する。Here, n is a natural number, the equation (1) is changed, and the pitch coefficient k is defined by the following equation (2).
【００２３】[0023]
【数２】 [Equation 2]
【００２４】この式（２）でピッチ係数ｋが整数となる
様に、レンズの倍率ｍと、自然数ｎを設定する。この整
数は、小さい方が撮像画像のパターンを消去しやすい。
一例として、被検査物のパターンのピッチＰが２２μ
ｍ、固体撮像素子（ＣＣＤ）の画素のピッチｃが１２μ
ｍ、レンズ倍率ｍを３、自然数ｎを２とすると、ピッチ
係数ｋは１１となり整数となる。この場合の濃度プロフ
ァイルは図３のｆ７に示す様な形状になる。つまり、自
然数ｎが２となっているため、ｎ×ｐつまり２×ｐ間離
れたパターン４２の濃度プロファイルｆ７は同じ形にな
る。こうして、濃度プロファイルｆ７に対しパターンの
ピッチが２（ｎ）周期離れた濃淡値の差分をとり、被検
査物を撮像した画像のパターンを消去する。In the equation (2), the lens magnification m and the natural number n are set so that the pitch coefficient k becomes an integer. The smaller this integer is, the easier the pattern of the captured image is erased.
As an example, the pattern pitch P of the inspection object is 22μ.
m, the pixel pitch c of the solid-state image sensor (CCD) is 12μ
When m, the lens magnification m are 3 and the natural number n is 2, the pitch coefficient k becomes 11 and becomes an integer. The density profile in this case has a shape as shown by f7 in FIG. That is, since the natural number n is 2, n × p, that is, the density profiles f7 of the patterns 42 separated by 2 × p have the same shape. In this way, the difference between the grayscale values with the pattern pitch separated by 2 (n) cycles from the density profile f7 is obtained, and the pattern of the image of the object to be inspected is erased.
【００２５】つまり、式（２）でピッチ係数ｋが整数と
なる様に、レンズ倍率ｍと自然数ｎを決定し、撮像画像
のｎ周期離れたパターンの差分をとれば、撮像画像のパ
ターンを消去することが出来る。また、ｎの整数倍の周
期離れたパターンの差分をとっても、同様に撮像画像の
パターンを消去することが出来る。したがって、自然数
をｊとし、ｎ×ｊ周期離れた撮像画像のパターンの差分
をとってもよい。That is, the lens magnification m and the natural number n are determined so that the pitch coefficient k is an integer in the equation (2), and the pattern of the captured image is erased by taking the difference between the patterns of the captured image separated by n cycles. You can do it. Further, the pattern of the captured image can be similarly deleted by taking the difference between the patterns separated by a cycle that is an integral multiple of n. Therefore, the natural number may be j, and the difference between the patterns of the captured images separated by n × j cycles may be taken.
【００２６】一方、被検査物のパターンのピッチＰと、
固体撮像素子（ＣＣＤ）の画素のピッチｃは定数であ
り、レンズ倍率ｍの変更にも限界がある。このため、自
然数ｎとレンズ倍率ｍを変更しても、ピッチ係数ｋが自
然数にならない場合がある。しかし、ピッチ係数ｋが整
数に近い値の場合は、ｎ×ｊ周期離れたパターンの差分
をとりパターンを消去しても問題はないことが実験にて
明らかになった。具体的には、ピッチ係数ｋの小数点以
下を四捨五入した数をｉとし、ピッチ係数ｋが０．９×
ｉ≦ｋ≦１．１×ｉの範囲であれば、撮像画像のｎ×ｊ
周期離れたパターンの差分をとりパターンを消去するこ
とが出来る。一例として、被検査物のパターンのピッチ
Ｐが８８μｍ、固体撮像素子（ＣＣＤ）の画素のピッチ
ｃが１３μｍの場合、レンズ倍率ｍを４．９５、自然数
ｎを２とすると、ピッチ係数ｋは６７．０１５となり、
上記の条件０．９×ｉ≦ｋ≦１．１×ｉを満たす。On the other hand, the pattern pitch P of the object to be inspected,
The pixel pitch c of the solid-state image sensor (CCD) is a constant, and there is a limit to changing the lens magnification m. Therefore, the pitch coefficient k may not be a natural number even if the natural number n and the lens magnification m are changed. However, when the pitch coefficient k is a value close to an integer, it has been clarified by experiments that there is no problem even if the patterns are erased by taking the difference between the patterns separated by n × j periods. Specifically, the number obtained by rounding off the decimal point of the pitch coefficient k is i, and the pitch coefficient k is 0.9 ×
In the range of i ≦ k ≦ 1.1 × i, n × j of the captured image
The pattern can be erased by taking the difference between the patterns separated by a period. As an example, when the pattern pitch P of the object to be inspected is 88 μm and the pixel pitch c of the solid-state image sensor (CCD) is 13 μm, the pitch coefficient k is 67 when the lens magnification m is 4.95 and the natural number n is 2. .015,
The above condition 0.9 × i ≦ k ≦ 1.1 × i is satisfied.
【００２７】この様にして、被検査物の撮像画像からパ
ターンを消去した後、従来と同様の方法で、パターンを
消去した画像の濃度値が、濃度値が第１の所定の値以上
または第２の所定の値以下の部分を夫々欠陥とし検出す
る。In this way, after the pattern is erased from the imaged image of the object to be inspected, the density value of the image in which the pattern is erased is equal to or larger than the first predetermined value or the first value by the same method as the conventional method. The portions below the predetermined value of 2 are detected as defects.
【００２８】これにより、被検査物を固体撮像素子にて
撮像した時でも量子化誤差の少ない撮像画像を得ること
が可能になり、適切に撮像画像のパターンを消去し、正
確に欠陥を検査することが可能になる。As a result, it is possible to obtain a picked-up image with a small quantization error even when the object to be inspected is picked up by the solid-state image pickup element, the pattern of the picked-up image is appropriately erased, and the defect is accurately inspected. It will be possible.
【００２９】また、被検査物がプラズマディスプレイパ
ネルや、半導体ウェハであっても、同様の効果が得られ
る。Similar effects can be obtained even when the object to be inspected is a plasma display panel or a semiconductor wafer.
【００３０】[0030]
【発明の効果】以上の様に本発明によれば、固体撮像素
子の画素のピッチをｃとし、パターンのピッチをｐと
し、レンズの倍率をｍとし、自然数をｎとし、自然数を
ｊとし、ｋ＝ｎ×ｍ×ｐ／ｃで定義するピッチ係数ｋ
が、０．９×ｉ≦ｋ≦１．１×ｉの範囲になるようにレ
ンズの倍率ｍと自然数ｎを設定すれば、被検査物を固体
撮像素子にて撮像した時でも量子化誤差の少ない撮像画
像を得ることが出来る。As described above, according to the present invention, the pixel pitch of the solid-state image pickup device is c, the pattern pitch is p, the lens magnification is m, the natural number is n, and the natural number is j. pitch coefficient k defined by k = n × m × p / c
However, if the magnification m and the natural number n of the lens are set so that 0.9 × i ≦ k ≦ 1.1 × i, a quantization error of the object to be inspected by the solid-state image sensor will be obtained. It is possible to obtain a small number of captured images.
【００３１】この撮像画像のｎ×ｊ周期離れているパタ
ーンの差分をとり、被検査物を撮像した撮像画像のパタ
ーンを消去する事で、適切に撮像画像のパターンを消去
することができるので、白欠陥閾値と黒欠陥閾値の幅を
従来に比べて小さくすることができ、厳密な欠陥検査を
実施することが可能になる。By taking the difference between the patterns of the picked-up image separated by n × j cycles and deleting the pattern of the picked-up image of the object to be inspected, the pattern of the picked-up image can be appropriately deleted. The width between the white defect threshold and the black defect threshold can be made smaller than in the conventional case, and strict defect inspection can be performed.
【図１】本発明の実施の形態の撮像装置を示す概念図FIG. 1 is a conceptual diagram showing an image pickup apparatus according to an embodiment of the present invention.
【図２】本発明の実施の形態の撮像画像を量子化する概
念を示す図FIG. 2 is a diagram showing a concept of quantizing a captured image according to the embodiment of the present invention.
【図３】本発明の実施の形態のｎ＝２の時の撮像画像を
量子化する概念を示す図FIG. 3 is a diagram showing a concept of quantizing a captured image when n = 2 according to the embodiment of the present invention.
【図４】従来のパターン消去方法の原理を示す概念図FIG. 4 is a conceptual diagram showing the principle of a conventional pattern erasing method.
【図５】従来の量子化誤差が発生する原理を示す概念図FIG. 5 is a conceptual diagram showing a principle that a conventional quantization error occurs.
３１ 被検査物 ３２ レンズ ３３ カメラ ３４ ＣＣＤ（固体撮像素子） ４１ 撮像画像 ４２ パターン 31 inspected 32 lenses 33 cameras 34 CCD (Solid-state image sensor) 41 captured image 42 patterns
フロントページの続き (51)Int.Cl.7 識別記号 ＦＩ テーマコート゛(参考） Ｈ０４Ｎ 7/18 Ｈ０１Ｌ 27/14 Ｚ ５Ｃ０５４ (72)発明者 湯川 典昭 大阪府門真市大字門真1006番地 松下電器 産業株式会社内 (72)発明者 野村 剛 大阪府門真市大字門真1006番地 松下電器 産業株式会社内 Ｆターム(参考） 2G051 AA51 AB20 CA03 CA04 CB01 EA08 ED04 ED11 2H088 FA11 FA30 MA20 4M106 AA01 CA39 DB04 DB21 DJ11 DJ20 4M118 AA05 AB01 AB02 AB03 BA09 CA25 5B057 AA02 BA02 BA29 DA03 DB02 DB05 DB09 DC32 5C054 FC01 HA01 HA05 Front page continuation (51) Int.Cl. 7 identification code FI theme code (reference) H04N 7/18 H01L 27/14 Z 5C054 (72) Inventor Noriaki Yukawa 1006 Kadoma, Kadoma City, Osaka Matsushita Electric Industrial Co., Ltd. (72) Inventor Tsuyoshi Nomura, Osaka Prefecture Kadoma City, Oma 1006, Kadoma, F-Term, Matsushita Electric Industrial Co., Ltd. (reference) AB02 AB03 BA09 CA25 5B057 AA02 BA02 BA29 DA03 DB02 DB05 DB09 DC32 5C054 FC01 HA01 HA05
Claims (3)
ズを通して固体撮像素子にて撮像し、撮像画像の異なる
位置のパターンの濃度の差分をとり、撮像画像のパター
ンを消去するパターン消去方法において、 前記固体撮像素子の画素のピッチをｃとし、前記被検査
物のパターンのピッチをｐとし、前記レンズの倍率をｍ
とし、自然数をｎとし、自然数をｊとし、ｋ＝ｎ×ｍ×
ｐ／ｃの式でピッチ係数ｋを定義し、ピッチ係数ｋの小
数点以下を四捨五入した数をｉとし、ピッチ係数ｋが
０．９×ｉ≦ｋ≦１．１×ｉの範囲になるように前記レ
ンズの倍率ｍと自然数ｎを設定し、ｎ×ｊ周期離れてい
る前記撮像画像のパターンの差分をとり、前記撮像画像
全体のパターンを消去する事を特徴とするパターン消去
方法。1. A pattern erasing method for erasing a pattern of a picked-up image by picking up an image of an object to be inspected having a repetitive pattern with a solid-state image pickup element through a lens, obtaining a difference in density of patterns at different positions of the picked-up image, The pixel pitch of the solid-state image sensor is c, the pattern pitch of the object to be inspected is p, and the magnification of the lens is m.
, And the natural number is n, the natural number is j, and k = n × m ×
The pitch coefficient k is defined by the equation of p / c, and the number rounded off to the right of the decimal point of the pitch coefficient k is set to i, so that the pitch coefficient k falls within the range of 0.9 × i ≦ k ≦ 1.1 × i. A pattern erasing method, characterized in that a magnification m of the lens and a natural number n are set, a difference between patterns of the captured image separated by n × j periods is calculated, and the pattern of the entire captured image is erased.
い、前記パターンを消去した撮像画像の濃度値が第１の
所定の値以上または第２の所定の値以下の部分を夫々欠
陥とする事を特徴とする欠陥検査方法。2. The pattern erasing method according to claim 1, wherein a portion where the density value of the captured image from which the pattern has been erased is equal to or more than a first predetermined value or less than a second predetermined value is determined as a defect. Defect inspection method characterized by things.
液晶パネル、プラズマディスプレイパネル、あるいは半
導体ウェハの欠陥を検査する事を特徴とする欠陥検査方
法。3. The defect inspection method according to claim 2,
A defect inspection method characterized by inspecting a liquid crystal panel, a plasma display panel, or a semiconductor wafer for defects.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2002134189A JP4165115B2 (en) | 2002-05-09 | 2002-05-09 | Pattern erasing method and defect inspection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2002134189A JP4165115B2 (en) | 2002-05-09 | 2002-05-09 | Pattern erasing method and defect inspection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP2003329609A true JP2003329609A (en) | 2003-11-19 |
| JP4165115B2 JP4165115B2 (en) | 2008-10-15 |
Family
ID=29696916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2002134189A Expired - Fee Related JP4165115B2 (en) | 2002-05-09 | 2002-05-09 | Pattern erasing method and defect inspection method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP4165115B2 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005346650A (en) * | 2004-06-07 | 2005-12-15 | Sony Corp | Image processing system, image processor, image processing method and program |
| CN101923060A (en) * | 2009-06-04 | 2010-12-22 | 松下电器产业株式会社 | Shooting check method |
| JP2011215132A (en) * | 2010-03-16 | 2011-10-27 | Panasonic Corp | Pattern inspection method, pattern inspection device, and imaging head for the pattern inspection device |
| KR101217906B1 (en) * | 2004-06-07 | 2013-01-02 | 소니 주식회사 | Image processing apparatus and method, and recording medium |
| JP2018155600A (en) * | 2017-03-17 | 2018-10-04 | 東レエンジニアリング株式会社 | Visual inspection device |
| JP2020136368A (en) * | 2019-02-15 | 2020-08-31 | 東京エレクトロン株式会社 | Image generation apparatus, inspection apparatus, and image generation method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102221562B (en) * | 2010-03-16 | 2013-12-25 | 松下电器产业株式会社 | Pattern checking method and pattern checking device |
-
2002
- 2002-05-09 JP JP2002134189A patent/JP4165115B2/en not_active Expired - Fee Related
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005346650A (en) * | 2004-06-07 | 2005-12-15 | Sony Corp | Image processing system, image processor, image processing method and program |
| JP4617726B2 (en) * | 2004-06-07 | 2011-01-26 | ソニー株式会社 | Image processing system, image processing apparatus, image processing method, and program |
| KR101217906B1 (en) * | 2004-06-07 | 2013-01-02 | 소니 주식회사 | Image processing apparatus and method, and recording medium |
| CN101923060A (en) * | 2009-06-04 | 2010-12-22 | 松下电器产业株式会社 | Shooting check method |
| JP2011215132A (en) * | 2010-03-16 | 2011-10-27 | Panasonic Corp | Pattern inspection method, pattern inspection device, and imaging head for the pattern inspection device |
| JP2018155600A (en) * | 2017-03-17 | 2018-10-04 | 東レエンジニアリング株式会社 | Visual inspection device |
| JP2020136368A (en) * | 2019-02-15 | 2020-08-31 | 東京エレクトロン株式会社 | Image generation apparatus, inspection apparatus, and image generation method |
| JP7166189B2 (en) | 2019-02-15 | 2022-11-07 | 東京エレクトロン株式会社 | Image generation device, inspection device and image generation method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP4165115B2 (en) | 2008-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6615172B2 (en) | Systems, devices and methods for quality assessment of OLED stack films | |
| CN102907083B (en) | Camera head, image processing apparatus and image processing method | |
| KR101373357B1 (en) | Camera exposure optimization techniques that take camera and scene motion into account | |
| US7995097B2 (en) | Techniques of motion estimation when acquiring an image of a scene that may be illuminated with a time varying luminance | |
| CN101339348B (en) | Image sensing apparatus and image sensing method | |
| CN1867042A (en) | System and method for subtracting dark noise from an image using an estimated dark noise scale factor | |
| JP2005158780A (en) | Method and device for inspecting defect of pattern | |
| DE602004031610D1 (en) | METHOD AND DEVICE FOR REDUCING THE EFFECTS OF DARK CURRENT AND DEFECTIVE PIXELS IN A PICTURE DEVICE | |
| JP2009222624A (en) | Sample inspection device and sample inspection method | |
| JP2003244521A (en) | Information processing method and apparatus, and recording medium | |
| JP2004012325A (en) | Method and apparatus for inspection of defect | |
| CN1917587A (en) | Image pick-up device and image processing method | |
| CN1452386A (en) | Camera unit | |
| JP2006050337A (en) | Imaging apparatus, imaging method, and imaging control program | |
| KR20180015139A (en) | Optical device for detecting internal defects of transparent substrate and method therefor | |
| JP2003329609A (en) | Method of deleting pattern and method of inspecting defect | |
| JP2008011467A (en) | Imaging method and apparatus for display panel | |
| US9179060B2 (en) | Method and apparatus for camera shake effect image stabilization | |
| JP2014115245A (en) | Defect inspection method for substrate, defect inspection device for substrate, program, and computer storage medium | |
| JP5108696B2 (en) | Imaging device | |
| US20130076940A1 (en) | Photographing apparatus and method of detecting distortion thereof | |
| CN104038689B (en) | Image processing apparatus and image processing method | |
| JPH1114334A (en) | Defect detecting method and device of netset or wired glass | |
| CN1470932A (en) | Optical apparatus | |
| JP2008032602A (en) | Surface defect inspecting apparatus, surface defect inspection method and method for manufacturing circuit board using the inspection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20050406 |
|
| RD01 | Notification of change of attorney |
Free format text: JAPANESE INTERMEDIATE CODE: A7421 Effective date: 20050706 |
|
| A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20070830 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20070911 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20071107 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20080708 |
|
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20080721 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20110808 Year of fee payment: 3 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20110808 Year of fee payment: 3 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20120808 Year of fee payment: 4 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20130808 Year of fee payment: 5 |
|
| S533 | Written request for registration of change of name |
Free format text: JAPANESE INTERMEDIATE CODE: R313533 |
|
| R350 | Written notification of registration of transfer |
Free format text: JAPANESE INTERMEDIATE CODE: R350 |
|
| S111 | Request for change of ownership or part of ownership |
Free format text: JAPANESE INTERMEDIATE CODE: R313113 |
|
| R350 | Written notification of registration of transfer |
Free format text: JAPANESE INTERMEDIATE CODE: R350 |
|
| LAPS | Cancellation because of no payment of annual fees |