JP2004219537A

JP2004219537A - Confocal microscope

Info

Publication number: JP2004219537A
Application number: JP2003004285A
Authority: JP
Inventors: Yuichi Tamura; 雄一田村
Original assignee: Nikon Corp; Nikon Engineering Co Ltd
Current assignee: Nikon Corp; Nikon Engineering Co Ltd
Priority date: 2003-01-10
Filing date: 2003-01-10
Publication date: 2004-08-05
Anticipated expiration: 2023-01-10
Also published as: JP4426763B2

Abstract

<P>PROBLEM TO BE SOLVED: To provide a confocal microscope which is made smaller in size than an imaging microscope. <P>SOLUTION: The confocal microscope is provided with: a light emitting means 21 wherein a plurality of light emitting pixels capable of emitting light are two-dimensionally arranged; a light receiving means 32 having a plurality of light receiving pixels corresponding to the light emitting pixels of the light emitting means 21; a confocal optical system for irradiating a sample 25 with the light emitted from the light emitting means 21 and condensing the return light to the light receiving means 32; and a control means 40 for setting one or adjacent light emitting pixels among the plurality of light emitting pixels as functional light emitting pixels, setting the functional light emitting pixels in several positions not adjacent to each other, setting the light receiving pixels corresponding to the plurality of functional light emitting pixels as functional light receiving pixels, and performing processing of emitting the light from the functional light emitting pixels and receiving the return light by the functional light receiving pixels a plurality of times for the functional light emitting pixels set by the different light emitting pixels among the plurality of light emitting pixels or set by the different combination of the light emitting pixels so as to obtain the image of the sample 25. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【０００１】
【発明の属する技術分野】
本発明は、共焦点顕微鏡に関する。
【０００２】
【従来の技術】
共焦点顕微鏡は従来の所謂結像型顕微鏡に比較して、面内分解能や軸上分解能（焦点深度）が２〜３割向上し、且つ試料のピンボケ像ノイズや光学系のフレアー等が無く、高コントラストで３次元画像が容易に得られるため、多くの分野（特に工業分野や生物・蛍光観察）で広く使用されるようになっている。
【０００３】
従来型共焦点顕微鏡の代表例には、回転するニポウディスクで光ビームを機械的に走査し、２次元検出器で受光する方式がある（例えば特許文献１参照）。
【０００４】
この方式では、回転するニポウディスク上に形成された透光用の複数のピンホールが所定間隔で螺旋状に配列されている。光源から出射された光が反射鏡とレンズにより構成される光学系で平行光線となり、そのピンホールの一つに照射される。照射光の一部はピンホールを透過後、対物レンズを透して試料に照射される。ニポウディスクの回転に伴ってピンホールがビームを横切り、ピンホール透過光は試料の弧状の領域を走査する。更に、ニポウディスクが回転することで、螺旋状に配置されたピンホールを透過した光が、次々と試料の別の領域を弧状に走査する。その結果、ピンホール透過光で試料を２次元走査することができる。その試料からの反射光を２次元検出器等で検出する。
【０００５】
【特許文献１】
特開平９−３２９７４８号公報「共焦点顕微鏡」
【０００６】
【発明が解決しようとする課題】
しかしながら、上記のような従来の共焦点顕微鏡は、結像型顕微鏡と比べて、機械的な光ビーム走査系が必要となるため、装置が大型化するという問題点があった。
【０００７】
この発明はこのような従来の問題点に鑑みてなされたもので、小型化が可能な共焦点顕微鏡を提供することを目的とする。
【０００８】
【課題を解決するための手段】
上記課題の解決のため、請求項１に係る発明は、
光を出射することが可能な複数の出射画素が、２次元的に配置された出射手段と、
前記出射手段の前記複数の出射画素に対応する複数の受光画素を有する受光手段と、
前記出射手段から出射した光を試料に照射し、前記試料面からの戻り光を前記受光手段に集光する共焦点光学系と、
前記複数の出射画素のうちの１つ又は隣合う複数の画素を機能出射画素として設定し、前記機能出射画素を隣合わない位置に少なくとも２つ設定し、前記複数の受光画素のうち、複数の前記機能出射画素に対応する受光画素を機能受光画素として設定するとともに、前記機能出射画素から光を出射させ前記機能受光画素で前記戻り光を受光させる処理を、前記複数の出射画素のうちの異なる出射画素あるいは異なる組み合わせの出射画素で設定された機能出射画素について複数回行うことにより前記試料の像を得る制御手段とを備えることを特徴とする。
【０００９】
請求項２に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記少なくとも２つの機能出射画素の間隔及び前記少なくとも２つの機能出射画素に対応する少なくとも２つの機能受光画素の間隔が、それぞれのエアリーディスク径の１０倍を越え１００倍以内であることを特徴とする。
【００１０】
請求項３に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、前記少なくとも２つの機能出射画素の間隔及び前記少なくとも２つの機能出射画素に対応する少なくとも２つの機能受光画素の間隔を、前記試料面の凹凸の段差が大きいほど大きく設定することを特徴とする。
【００１１】
請求項４に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、前記少なくとも２つの機能出射画素の間隔及び前記少なくとも２つの機能出射画素に対応する少なくとも２つの機能受光画素の画素単位の間隔を、
（試料の凹凸の段差）／（２．４４ｘ焦点深度）
より大きく設定することを特徴とする。
【００１２】
請求項５に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、使用対物レンズの瞳径に依存するエアリーディスク径に応じて、前記機能出射画素及び前記機能受光画素を構成する画素数を変化させることを特徴とする。
請求項６に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、前記機能出射画素から光を出射させ前記機能受光画素で前記戻り光を受光させる処理を、前記複数の出射画素のうちの異なる組み合わせの出射画素で設定された機能出射画素について複数回行う際に、複数回設定される前記機能出射画素どうしの中心間の距離を、１出射画素分あるいは前記機能出射画素を構成している縦方向あるいは横方向の出射画素数分以下に設定することを特徴とする。
【００１３】
請求項７に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、複数の前記機能出射画素及び前記受光機能画素の配置を正方配置又は六方配置又は全有効画面の縦横比に等しい比率の矩形配置又は全有効画面の縦横比に等しい比率の交互配置とすることを特徴とする。
【００１４】
請求項８に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記制御手段は、前記機能出射画素から光を出射させ前記機能受光画素で前記戻り光を受光させる処理を、前記複数の出射画素のうちの異なる出射画素で設定された機能出射画素について複数回行うことにより前記試料の像を得る際に、間引き走査及び補間により前記試料の画像を構成することを特徴とする。
【００１５】
請求項９に係る発明は、
請求項１記載の共焦点顕微鏡であって、
前記共焦点光学系は、
最大歪曲収差および最大倍率色収差を前記受光画素寸法の１／５以下に補正したコリメータレンズ、第１対物レンズ及び第２対物レンズとを有することを特徴とする。
【００１６】
請求項１０に係る発明は、
請求項９記載の共焦点顕微鏡であって、
前記第１対物レンズ及び第２対物レンズは、最大主光線傾斜角が焦点深度内でのずれ量を前記受光画素の寸法の１／５以内になるような角度に補正してあることを特徴とする。
【００１７】
【発明の実施の形態】
以下に、本発明の一実施形態について説明する。
【００１８】
図１は、本発明の一実施形態に係る共焦点顕微鏡の概略構成図である。
【００１９】
この共焦点顕微鏡は、光源１１、単色フィルタ１２及び集光レンズ１３を有する照明系１と、２次元シャッタアレイ２１、コリメータレンズ２２、ビームスプリッタ２３及び無限遠系対物レンズ２４（有限系対物レンズも構成可能である）を有する送光系２と、第２対物レンズ３１及び２次元検出器３２を有する受光系３と、２次元シャッタアレイ２１及び２次元検出器３２を制御する制御装置４０を有する制御系４と、を備えている。
【００２０】
ここで、２次元シャッタアレイ２１はコリメータレンズ２２の焦点位置にある。２次元シャッタアレイ２１は、本実施形態では、２次元のデジタルマイクロミラーデバイス（ＤＭＤ）で構成されており、照明系１からの光を反射して出射する複数の画素を有する。２次元シャッタアレイ２１の画素を出射画素と呼ぶ。また、その１つ乃至複数の連続した出射画素が光を、コリメータレンズ２２に出射している場合に、その出射画素を機能出射画素２１１と呼び、光を出射していない場合を遊休出射画素２１２と呼ぶことにする。
【００２１】
２次元シャッタアレイ２１を高速２次元液晶シャッタ（透過系）等に代替することも可能である。
【００２２】
光源１１としてレーザや通常の高輝度光源（例えば、水銀灯、メタルハライド）等が使用可能である。なお、２次元シャッタアレイ２１に２次元デジタルマイクロミラーデバイス（ＤＭＤ）を利用した場合、光源１１にＤＭＤに同期するパルス発信のレーザやＬＥＤのアレイを利用してもよい。この場合には、ＤＭＤのミラーパターン変更に合わせてパルス状の光信号を間欠的に送ればよく（機能画素２１１に照射する）、照射を続ける他の光源と比べて、光利用効率が向上する。
【００２３】
なお、２次元レーザダイオード（ＬＤ）アレイを使用する場合は、２次元シャッタアレイ２１と照明系１を省略でき、より小型・軽量・効率的な装置になる。
【００２４】
２次元検出器３２は、第２対物レンズ３１の焦点位置にある。２次元検出器３２は、２次元シャタアレイ２１内の複数の出射画素に対応する複数の受光画素を有する。受光画素の受光機能が働いている１つ乃至複数の連続した画素を機能受光画素３２１呼び、受光機能が働いていない画素を遊休受光画素３２２と呼ぶこととする。
【００２５】
コリメータレンズ２２、ビームスプリッタ２３、対物レンズ２４、及び第２対物レンズ３１からなる光学系は、２次元シャタアレイ２１の機能出射画素２１１、試料２５、及び２次元検出器３２上の機能受光画素３２１に対して、共焦点光学系となっている。すなわち、共焦点光学系は照明系がコリメータレンズ２２、結像系が第２対物レンズ３１で構成され、ビームスプリッタ２３と対物レンズ２４とは照明系、結像系で共用され、対物レンズ２４の焦点位置に試料を配置する。
【００２６】
次に、上記構成の共焦点顕微鏡の動作を説明する。
【００２７】
光源１１から出射した光は、単色フィルタ１２を透過後、集光レンズ１３により、２次元シャッタアレイ２１に集光される。２次元シャッタアレイ２１上の機能出射画素２１１で反射された光は、コリメータレンズ２２によって平行光線へと変換され、ビームスプリッタ２３を透過した後、対物レンズ２４に至る。対物レンズ２４に入射した光は試料２５に照射される。
【００２８】
試料２５からの反射光は対物レンズ２４により平行光線となり、ビームスプリッタ２３で反射された後、第２対物レンズ３１に到達する。試料２５からの反射光は、２次元検出器３２の機能受光画素３２１によって、機能出射画素２１１に対応する共焦点画像として検出される。本実施例の共焦点顕微鏡において、機能出射画素は出射共焦点ピンホール、機能受光画素は受光共焦点ピンホールとして機能している。以下、本明細書では、単に「機能画素」と記した場合は、機能出射画素、機能受光画素の両方を指すものとする。
【００２９】
試料２５全体の画像を得るために、制御装置４０が搭載された制御用ソフトウエアを通して、共役な位置にある機能画素２１１及び３２１に対して、以下のような操作を行なう。
【００３０】
制御装置４０は、２次元シャッタアレイ２１内の出射画素のうちの複数の画素を機能させ、送光系２に対して送光する。一方、制御装置４０は、２次元検出器３２内の受光画素から、送光した２次元シャッタアレイ２１内の複数の画素に対応する受光画素を選ぶ。そして、選ばれた複数の受光画素を機能させて受光させ、複数の出射画素に対応する試料２５の像を得る。したがって、一回の走査で、複数点の試料２５に対する共焦点画像情報が得られる。この処理（同時多点処理と呼ぶ）を全ての画素を覆うまで複数回行ない、試料２５全体の像を得る。
【００３１】
上記構成で全出射画素に対応する共焦点画像を得るためには、まず、２次元シャッタアレイ２１及び２次元検出器３２の使用領域及び画素形状は互いに相似形で且つ使用領域内同一画素数となっていることが好ましい。更に２次元シャッタアレイ２１の各出射画素からの出射光が２次元検出器３２の対応する各受光画素に入射する様に、コリメータレンズ２２の焦点距離すなわち倍率と、第２対物レンズ３１の焦点距離即ち倍率とを決定する。
【００３２】
２次元シャッタアレイ２１の最適な機能画素寸法（画素に内接する円の直径即ち最適共焦点ピンホール径）は、良く知られている様に、像が高解像度、高感度、高効率となるために、エアリーディスク径と略同一とすることが望ましい。画素寸法がエアリーディスク径を越えると解像が低下し、エアリーディスク径より小さいと感度や効率が低下するからである。そこでこの最適な画素寸法になるように、逆に対物レンズ２４の物体側開口数（ＮＡ）と焦点距離とから、コリメータレンズ２２の焦点距離すなわち倍率を決定する。
【００３３】
上記のような構成によって、大型高価な機械的なビーム走査機構を除去し、量産効果による安価な２次元シャッタアレイ（例えば、ＤＭＤ）２１や２次元検出器３２を使用することにより、装置全体を小型・軽量・安価にできる。また、多点走査方式の撮像により、ＴＶ（ビデオ）レート（毎秒３０画像）以上の高速走査・撮像を実現する。そして機械的なビーム走査機構が不用のため、直線や回転走査に伴うノイズ、走査光強度むらや画像歪が無く、低倍率・高開口数対物レンズの使用も可能になる。更に、最適共焦点ピンホール径を制御装置４０に備わる制御用ソフトウエアにより迅速に変更可能なため、瞳径（＝２ｘ焦点距離ｘ開口数）が異なり従ってエアリーディスク径の異なる種々の対物レンズの使用が可能で使用範囲が大幅に拡大する、という効果がある。
【００３４】
図２は、２次元シャッタアレイ２１と２次元検出器３２との同期多点走査の際に使われるＭ個の機能画素パターン（１〜Ｍ）を、時間軸に沿って並べた図である。以下では機能出射画素と機能受光画素は常に同一形状・配置で同期して機能するため、まとめて機能画素と呼ぶことにする。
【００３５】
多点走査では、（総画素数／機能画素数）の機能画素パターン数（Ｍ）がある。１画像全体は、多点走査を機能画素パターン数だけ繰り返すことにより得られる。例えば、総画素数が１００万画素の場合、機能画素数が１万画素であれば、１００回の走査で１画像を得られる。多点走査の機能画素の等間隔配列は機能画素単位の機能画素間隔を横方向、縦方向をそれぞれＰｘ、Ｐｙ＝機能画素間隔／機能画素（自然数）とし、縦列のずれ量をＤ＝縦列のずれ／機能画素（ゼロ、自然数）として表示できる（図７参照）。１画像を構成する機能画素パターン数はＰｘ・Ｐｙであり、機能画素数は総画素数／機能画素パターン数（Ｍ）となる。全画面は１つの機能画素を含むＰｘ・Ｐｙの走査範囲の繰り返しとなるため、これを単位走査範囲と呼ぶ。縦列のずれＤは縦列の単位走査範囲の横ずれでもある。（図３〜６参照）種々の機能画素配列のうち、便宜的にＰｘ＝Ｐｙ＝Ｐ、Ｄ＝０を正方配置と呼び（図３参照）、Ｐｘが偶数で、Ｐｙ＝Ｐｘ・ｓｉｎ（π／３）〜最近自然数、Ｄ＝Ｐｘ／２を六方配置と呼ぶ（図４参照）。また全有効画面の正方画素の縦横方向の画素数をＴｙ，Ｔｘとすると、縦横比Ｔｙ：Ｔｘ＝Ｐｙ：Ｐｘ、Ｄ＝０を全有効画面比例の矩形配置と呼び（図５参照）、Ｔｙ：Ｔｘ＝Ｐｙ：Ｐｘ，Ｄ＝Ｐｘ／２〜最近自然数を全有効画面比例の交互配置と呼ぶ（図６参照）事にする。
【００３６】
以下図３〜６に、本実施形態に係る共焦点顕微鏡の、全有効画面形状、機能画素の間隔及び配置と走査例を示す。以下の例では簡単のため少ない画素数で説明するが、現実の装置構成では２次元シャッタアレイと２次元検出器の画素数は通常Ｔｘ〜Ｔｙ〜１０００程度であり、機能画素間隔もＰｘ，Ｐｙ＝１０〜１００程度である。
【００３７】
図３は、機能画素（共焦点ピンホール）パターンと走査方式の具体例を、時間の経過と共に表示している。この図では全有効画面Ｔｘ・Ｔｙ＝１２ｘ１２画素、正方配置Ｐｘ＝Ｐｙ、機能画素間隔Ｐ＝６、Ｄ＝０の例で機能画素数は４である。ＰｘＰ＝３６パターンの走査で１画像が取得できる。多点走査では相対的な機能画素の配置を保ったまま時間と共に機能画素の位置をずらして撮像する。細い実線で示される部分は１つの機能画素を含む単位走査範囲で、全ての機能画素はそれぞれの単位走査範囲内を１回走査する事により１画像が構成される。３Ａは１画素ずらしながら走査する順次走査、３Ｂは２画素ずらしならが走査し、１画素ずらして再度走査する飛び越し走査、３Ｃは２画素ずらしながら走査し、不足画像を補間処理する間引き走査、３Ｄは斜めに１画素ずらしながら走査し、不足画像を補間する間引き走査を示している。特に、間引き走査は動態の高速撮像や３次元画像の撮像等の時に有効である。即ち、先ず１〜数画素の間引き撮像を行い、後にまとめて補間処理で画像を完成すれば、順次走査より２〜数倍の撮像速度が実現できる。
【００３８】
図４は全画面Ｔｘ・Ｔｙ＝１２ｘ１２画素、六方配置、機能画素間隔Ｐ＝６、Ｄ＝３の例である。全有効画面の面積が細い実線で示される単位走査範囲の面積の倍数でないため機能画素数が走査途中で４〜６と変化する。１画像はＰｘ・Ｐｙ＝ＰｘＰ・ｓｉｎ（π／３）〜６ｘ５＝３０機能画素パターンで構成される。４Ａ〜４Ｄは順次走査、１画素飛び越し走査、間引き走査を示している。六方配置は同一間隔の正方配置より像の対称性が良く（後述）、また機能画素パターン数が少なく高速撮像が可能である。
【００３９】
図５は全有効画面Ｔｙ・Ｔｘ＝９ｘ１２画素、全画面比例の縦横比Ｐｙ：Ｐｘ＝３：４、Ｄ＝０の矩形配置の例で、機能画素数は９ｘ１２／３ｘ４＝９である。１画像はＰｘ・Ｐｙ＝１２機能画素パターンで構成される。５Ａ〜５Ｄは順次走査、１画素飛び越し走査、間引き走査を示している。全有効画面比例の矩形配置は画像取得・処理が容易であるが、画像の対称性は低い（後述）。
【００４０】
図６は全有効画面Ｔｙ・Ｔｘ＝９ｘ１２画素、全画面比例の縦横比Ｐｙ：Ｐｘ＝３：４、Ｄ＝Ｐｘ／２＝２の交互配置の例で、機能画素は９ｘ１２／３ｘ４＝９である。１画像はＰｘ・Ｐｙ＝１２の走査パターンで構成される。６Ａ〜６Ｄは順次走査、１画素飛び越し走査、間引き走査を示している。全画面比例の配置のため画像取得・処理が容易であり、同一間隔の矩形配置に比較して画像の対称性が高い（後述）。
【００４１】
以下図７〜１１に、本実施形態に係る共焦点顕微鏡の、出射画素の寸法、機能出射画素の間隔、配置例及び特徴を示す。
【００４２】
多点走査では同時に多数の機能画素が存在し、試料凹凸によるそれらのピンボケ像がノイズとなる（後述）。機能画素間の距離が方向により異なると像のノイズ即ちコントラストに方向性が出る。この像の対称性は次式Ｓで概略評価できる。
【００４３】
Ｓ＝最小・最大コントラスト比＝最大・最小ノイズ比＝（最小間隔／最大間隔）^２正方配置ではＳ＝Ｐ^２／（Ｐ^２＋Ｐ^２）＝０．５、六方配置では全方向の間隔は一定でＳ＝｛（Ｐｘ／２）^２＋（Ｐｘｓｉｎ（π／３））^２）／Ｐｘ^２＝１従って六方配置は方向によるコントラストの違いの無い対称性の良い配置である。像の対称性の影響は機能画素間隔Ｐが小さく、低コントラスト試料の観察の場合に大きいが、逆の場合は無視できる。
【００４４】
図７は全有効画面Ｔｙ・Ｔｘ＝２７ｘ４８画素、ハイビジョン方式の全有効画面相似の交互配置の例であり、実用上の効果が期待できる。縦横比はＴｙ：Ｔｘ＝Ｐｙ：Ｐｘ＝９：１６、Ｄ＝Ｐｘ／２＝８、機能画素数は２７ｘ４８／９ｘ１６＝９である。１画像はＰｘ・Ｐｙ＝１４４の機能画素パターンで構成される。図中の機能画素間の２つの矢印と数式は最短・最長の機能画素（共焦点ピンホール）間隔を示しており、像の対称性はＳ＝（Ｄ^２＋Ｐｙ^２）／Ｐｘ^２〜０．５７となる。
【００４５】
図８は２ｘ２画素からなる機能画素でＴＶやビデオの全有効画面（Ｔｙ・Ｔｘ＝３６ｘ４８）相似の３：４矩形配置（Ｐｙ＝６，Ｐｘ＝８，Ｄ＝０）の例であり、実用上の効果を期待できる。機能画素（共焦点ピンホール径）は２ｘ２画素構成で中大瞳径の対物レンズ使用に適している。矩形配置のため像の対称性はＳ＝Ｐｙ^２／（Ｐｘ^２＋Ｐｙ^２）＝０．３６で低い。
【００４６】
多点走査では１機能画素が１画素で構成される場合は１機能画素（１画素）ピッチで走査するが、１機能画素が２ｘ２画素構成のように複数画素から構成される場合には、走査ピッチを１画素から１機能画素（２画素）まで可変制御できる。１機能画素ピッチより細かいピッチの走査では解像は向上するが、走査パターン数はＰｘ・Ｐｙより多くなり、撮像時間も長くなる。通常は良好な解像を達成するために最小エアリーディスク径に相当する機能画素を２ｘ２画素構成とし、１画素ピッチで走査する。この場合１機能画素（２画素）ピッチの走査に比較して解像は２倍程度になるが、走査パターン数は４倍、撮像時間も４倍になる。この機能画素の走査ピッチは目的や試料によりソフトウエアで容易に選択・制御可能である。
【００４７】
一般に共焦点ピンホール（機能画素）の最適径はエアリーディスク径程度である（既述）。波長をλ、屈折率をｎ、対物レンズの開口角をθ、開口数をＮＡ＝ｎｓｉｎθ、倍率をβとすると、像側開口数はＮＡ’＝ｎ’ｓｉｎθ’＝ＮＡ／βとなり、像側（機能画素側）エアリーディスク径は２δ＝１．２２λ／ＮＡ’＝１．２２λｘβ／ＮＡとなる。倍率や種別の異なる対物レンズを使用すると像側開口数（瞳径）が大きく変化するため、エアリーディス径も変化する。対物レンズの倍率と開口数を倍率ｘ／開口数と表示すると、現実の対物レンズは５ｘ／０．２５、１０ｘ／０．３、１００ｘ／０．９等であり、それぞれＮＡ’＝０．０５、０．０３、０．００９となり、エアリーディスク径も５倍規準でそれぞれ１、３．３、５．６程度と大きく異なる。従って多種類の対物レンズを使用する場合、最適共焦点ピンホール径をそのエアリーディスク径にあわせるため１つの機能画素を構成する画素の数を増減させる必要がある。具体的には５ｘ／０．２５の対物レンズの機能画素を２ｘ２画素構成とした場合（走査ピッチは１画素）、１０ｘ／０．３の機能画素は３ｘ３画素構成、１００ｘ／０．９の機能画素は１１ｘ１１画素構成の様にする。
【００４８】
特に、従来像側開口数が小さい対物レンズと同時使用が困難であった低倍率・中開口数対物レンズにより、試料の広範囲を高解像力で観察・計測でき、効率が向上する。即ち、同じ開口数で倍率が２倍違う対物レンズを比較すると、倍率が半分の対物レンズでは４倍広い範囲を観察・計測でき、効率が４倍になる。
【００４９】
プリント配線板、回折光学素子やＭＥＭＳ（ＭｉｃｒｏＥｌｅｃｔｒｏＭｅｃｈａｎｉｃａｌＳｙｓｔｅｍ）等の微小部品の３次元形状検査では、共焦点顕微鏡により手動で設計データとその３次元形状検査結果を比較して微小部品形状の良否を判定している。一般的な検査工程ではこの種の部品は形状・大きさが多種多様であるため、部品の種類に応じて像側開口数の異なる多種の対物レンズに切り替えて使用する必要がある。しかし従来のニポウディスクによる方式では、ニポウディスクに形成されたピンホール径の変更は容易でない。従って従来型の共焦点顕微鏡では限定された範囲の対物レンズのみしか有効に使用できない。
【００５０】
本実施形態の共焦点顕微鏡では、機能画素の寸法（最適共焦点ピンホール径）を制御装置４０のソフトの制御により機能画素の構成画素数を増減するだけで変更・選択できるので、多種類の対物レンズ使用に対して迅速に対応できる。特に低倍率中開口数対物レンズを組み合わせることで、種々の微小部品の形状をこれまでより迅速且つ効率的に、しかもコントラストを制御して（後述）、検査することができる。更に、３次元ＣＡＤデータ化した設計データと、本発明の共焦点顕微鏡による３次元形状測定結果との比較を自動化することにより比較検査過程全体を迅速・効率化できる。
【００５１】
図９は５ｘ５画素構成で全有効画面Ｔｘ・Ｔｙ＝４５ｘ４５画素、正方配置（Ｐｘ＝Ｐｙ）の例である。機能画素間隔Ｐ＝５、Ｄ＝０より１画像はＰｘＰ＝２５機能画素パターンで構成される。像の対称性Ｓ＝Ｐ^２／（Ｐ^２＋Ｐ^２）＝０．５１００ｘ／０．９の対物レンズの様に瞳径の小さな対物レンズに適した例である。
【００５２】
図１０は３ｘ３画素構成で全有効画面Ｔｘ＝Ｔｙ＝４５ｘ４５画素、六方配置の例である。機能画素間隔Ｐｘ＝６、Ｐｙ＝Ｐｘ・ｓｉｎ（π／３）〜５（最近自然数）、Ｄ＝Ｐｘ／２＝３より１画像はＰｘ・Ｐｙ＝３０機能画素パターンで構成される。像の対称性Ｓ＝（Ｄ^２＋Ｐｙ^２）／Ｐｘ^２＝０．９４で理想的な六方配置の１より低いが実用上は問題が無い程度である。５０ｘ／０．８の対物レンズの様に瞳径の比較的小さな対物レンズに適した例である。
【００５３】
図１１は全有効画面Ｔｘ・Ｔｙ＝１５ｘ１５画素構成の場合の他の機能画素配列の例であり、上段は点配置即ち点走査共焦点顕微鏡で、撮像時間がかかるが最高コントラスト・最高画質が得られる。機能画素間隔はＰｘ＝Ｐｙ＝Ｐ＝Ｔｘ＝Ｔｙ＝１５である。中段は水平多線配置で、線走査共焦点顕微鏡より高速撮像が可能である。機能画素間隔はＰｘ＝１，Ｐｙ＝５であり、像の対称性Ｓ＝Ｐｘ^２／Ｐｙ^２＜０．０４は最低である。下段は全てを機能画素にした場合即ち結像型顕微鏡の場合で、最高の撮像速度が実現できるが像のコントラストは最低である。機能画素間隔はＰｘ＝Ｐｙ＝Ｐ＝１である。本実施形態においては、機能画素の配置はハードを変更せずに制御装置４０に備わる制御用ソフトウエアで目的や試料により、点、線、多点走査及び結像型（非光走査）などの走査方式の選択・制御が可能である。
【００５４】
多点走査では同時に多数の機能画素が存在するため、機能画素の間隔や配置により受光ノイズが増減する。配置による像の対称性は既述したが、以下では機能画素間隔とノイズの関連及び実用上の最適な機能画素間隔範囲について説明する。
【００５５】
図１２は、ある時間の２次元検出器の断面図であり、機能画素間隔とピンボケ像による光量クロストークの関係を図示している。厚い試料や凹凸の大きい試料の場合は、ピンボケ像による隣接機能画素に漏れる光量クロストークが現れる。ピンボケ像が隣接する機能画素にまで及ぶと光量ノイズとなり、コントラストが低下する。隣接している遊休画素への漏れは検出されないため無視できる。以下では簡単のため、結像型（Ｐ＝１）を基準に取り、機能画素は正方配置（Ｐｘ＝Ｐｙ＝Ｐ）の場合を考える。像側焦点深度はＤＯＦ’＝±λ／（２ＮＡ’^２）、実寸の機能画素間隔はＰｒ＝Ｐｘ１．２２λ／ＮＡ’、ｎ’＝１，ｔａｎθ’〜ｓｉｎθ’＝ＮＡ’、図１２より、隣接機能画素へ漏れないボケ量（光軸上距離）はη_Ｐ’＝Ｐｘ１．２２ｘλ／ＮＡ’^２＝Ｐｘ２．４４ｘＤＯＦ’となる。これに相当する試料凹凸（ピンボケノイズの無い試料凹凸量）をη_Ｐとすれば、η_Ｐ＝η_Ｐ’／β^２、ＮＡ＝β・ＮＡ’より、機能画素間隔Ｐに比例する次式が成立する。
【００５６】
η_Ｐ＝Ｐｘ１．２２ｘλ／ＮＡ^２＝Ｐｘ２．４４ｘＤＯＦ〜Ｐｘ２ｘＤＯＦ
例えば、結像型（Ｐ＝１）ではη_１＝１．２２ｘλ／ＮＡ^２＝２．４４ｘＤＯＦ、点走査共焦点（Ｐ＝Ｔｘ＝Ｔｙ）ではη_Ｔｘ＝Ｔｘ・２．４４ｘＤＯＦだが、機能画素がただ１つなので隣接画素からの漏れ光は無く凹凸に拠らず常にη_Ｔｘ＝∞となり、ピンボケノイズは無い。また、試料凹凸をηとし、η＞η_Ｐの場合、ピンボケノイズ（光量）は光量が保存する事からＰ^２に反比例する。更に、機能画素間隔Ｐを以下の様に制御すればピンボケノイズの無い撮像ができる。
【００５７】
Ｐ＞η／（２．４４ｘＤＯＦ）＝η’／（２．４４ｘＤＯＦ’）
具体的な凹凸が未知の試料の場合、最初は機能画素間隔Ｐを小さくして高速撮像をし、その後機能画素間隔Ｐを大きくして撮像するのが効率的である。
【００５８】
機能画素間隔Ｐと像のコントラストＣには、以下の関係がある。Ｉ_０を照射光量とすれば、ピンボケ像によるノイズはＮ_Ｆ∝Ｉ_０／Ｐ^２；η＞η_Ｐ、Ｎ_Ｆ＝０；η＜η_Ｐ光学系の反射によるフレアーノイズはＮ_Ｏ∝ Ｉ_０／Ｐ^２、総合した像のノイズは次式になる。
【００５９】
Ｎ＝Ｎ_Ｆ＋Ｎ_Ｏ∝ Ｉ_０／Ｐ^２
像のコントラストＣは、像強度をＩ∝Ｉ_０とすれば、Ｎ／Ｉ＜＜１の場合、次式になる。
【００６０】
Ｃ＝（Ｉ−Ｎ）／（Ｉ＋Ｎ）∝１−２Ｎ／Ｉ∝１−２／Ｐ^２
すなわち、機能画素間隔Ｐが大きい（機能画素の数が少ない）ほど、多点走査共焦点光学系のノイズやフレアーも減少し、コントラストＣは増加することになる。
【００６１】
一方、撮像速度Ｖはパターン数Ｐ^２に反比例するため、次式になる。
【００６２】
Ｖ∝Ｉ_０／Ｐ^２
光利用効率Ｅは、多点走査方式の場合には２次元シャッタアレイ全体に一様に光を照射し、機能出射画素のみ光を利用するため、次式になる。
【００６３】
Ｅ∝１／Ｐ^２
結像型（Ｐ＝１）ではコントラストＣは最小であり、撮像速度Ｖと光利用効率Ｅは最大である。Ｐの増加に伴ってコントラストは増加していくが、撮像速度と光利用効率は２乗で減少する。以上より、ピンボケノイズの無い試料凹凸、像のコントラストと撮像速度、光利用効率のバランスを取ることが好ましい。
【００６４】
具体例として、Ｐ＝１０（１０ｘ１０＝１００画素当たり１個を機能画素として使用の場合）、２０ｘ／０．７５対物レンズ、可視光の中心波長λ＝５５０ｎｍの場合、上式より無ノイズ凹凸はη〜０．０１２ｍｍとなり、焦点深度（凹凸の垂直分解能）ＤＯＦ＝±λ／（２ＮＡ^２）＝０．４９μｍの２４倍程度、空間分解能δ＝０．６１λ／ＮＡ＝０．４５μｍの２７倍程度である。無ノイズ凹凸を超える凹凸の試料の場合でも、結像型（Ｐ＝１）に比較したピンボケノイズは１／Ｐ^２＝１／１００になり、隣接する機能画素にもれる光量は僅かで十分共焦点光学系として機能する。また撮像時間も長くなく実用的である。
【００６５】
一方、Ｐ＝１００（１００ｘ１００＝１００００画素当たり１個を機能画素として使用の場合）では、同じ２０ｘ／０．７５の対物レンズ、波長λ＝５５０ｎｍを利用した場合に、無ノイズ凹凸はη〜０．１２ｍｍとＰ＝１０の場合に比較して１０倍大きくなる。つまり像の分解能は同じだが、１０倍厚い試料凹凸までノイズ無く撮像できる。またこの厚さを超える凹凸の試料でも、ノイズは結像型に比べて１／１００００、Ｐ＝１０の場合に比較して１／１００となる。よって、極低コントラスト試料（反射率や透過率の差が僅かな試料）の撮像が可能になる。このようにＰ＝１００の場合に、像のコントラストは大幅に向上する。しかし、撮像時間がＰ＝１０の場合と比較して１００倍かかり、光利用効率もＰ＝１０の場合と比較して１／１００程度と低くなる。このため最初に２次元や３次元試料の全体を高速に撮像し（Ｐ＝１０）、その後必要と思われる部分のみ再度高コントラスト・高画質のＰ＝１００で時間をかけて撮像する場合や、また予め低コントラスト試料と解っているものに対してＰ＝１００で撮像時間を掛けてノイズを低減した高コントラスト撮像する場合などが想定される。これ以上の機能画素間隔Ｐ＞１００では、更に高コントラスト撮像になるが撮像速度と光利用効率が一層低下し（＜１／１００００）、実用的でなくなる。もちろん本実施形態において、凹凸や厚さが既知の試料は上記機能画素間隔Ｐ＝１０〜１００の範囲を超えて、最適な機能画素間隔Ｐに変えることも、必要に応じて可能である。いずれの場合でも、機能画素間隔Ｐの変更は、制御装置４０に備わる制御用ソフトウエアにより容易に実効可能である。
【００６６】
以上より、目的や試料により機能画素間隔Ｐの値を変えることで、高速撮像（Ｐ＝１０：低コントラスト撮像を許容）、標準撮像（Ｐ〜３０）、高コントラスト撮像（Ｐ＝１００：低速撮像を許容）を選択することができる。また２次元や３次元試料で全体を先ず高速撮像（Ｐ＝１０）し、次に必要部分のみを高コントラスト・高画質撮像（Ｐ＝１００）を再度行うこと等もできる。従って、実用的な機能画素間隔はＰ＝１０〜１００程度で目的や試料により使い分ける事が適当である。
【００６７】
以上説明したように、本実施形態の多点走査共焦点顕微鏡では、ソフトウエアにより機能画素寸法（１つの機能画素を構成する画素数）で最適解像・効率と使用対物レンズの選択・制御が、走査ピッチで解像の選択・制御が、機能画素間隔で撮像速度、コントラスト（ノイズ）、光利用効率の制御が、機能画素配置で像の対称性の制御が、順次や間引き等の走査方法で撮像速度の制御が出来、目的や試料形態によりこれらを迅速・容易に変更・選択・使用が可能である。具体的には、機能画素の大きさをエアリーディスク径程度、構成画素数を２ｘ２、走査ピッチを１画素とし、その間隔をエアリーディスク径の１０〜１００倍程度としたことで、目的や試料により最適な撮像速度、光利用効率、高解像や高コントラストの画像が選択できる。またプリント配線板、回折光学素子やＭＥＭＳ（ＭｉｃｒｏＥｌｅｃｔｒｏＭｅｃｈａｎｉｃａｌＳｙｓｔｅｍ）等の微小部品の様に厚さや凹凸が既知の試料では機能画素間隔をＰ＞η／（２．４４ｘＤＯＦ）とすることにより、ピンボケ像ノイズをゼロにした高コントラストで効率的撮像ができる。そして、機能画素の配置を六方配置、正方（四方）配置、矩形配置又は交互配置としたり、間引き走査・補間処理をすることで対称性の良い画像の撮像やより迅速・効率的な撮像ができる。更に、機能画素間隔を大きく制御することにより、撮像速度の制御、点走査から多点走査を経て結像型顕微鏡まで任意の撮像方式が選択・使用できる。以上の説明では、まとめて機能画素の大きさや配置について説明したが、機能出射画素の大きさや配置を変えた場合、機能受光画素の大きさや配置もそれに伴って変えることは言うまでもない。
【００６８】
次に、図１３及び図１４で本実施形態の光学系に対する要件を説明する。本実施形態では、多点走査のため、画面全域に渡って光学像が満足すべき条件が加わる。２次元シャッタアレイ２１上の機能出射画素が２次元検出器３２上に形成する像と、２次元検出器３２上の対応する機能受光画素との位置がずれると受光量が低下し、ピンボケによる強度低下と区別がつかなくなり、取得画像の立体的な歪や像強度の忠実度に影響する。これを避けるため位置ずれの許容規準として、８０％以上の受光量確保のためにずれ面積が２０％以下にする事を要件とすると、像位置のずれは最大でも画素寸法の１／５以下に抑える必要がある。像位置のずれは、光学系のテレセントリック性による縦（光軸）方向ずれ（図１３）と歪曲収差、倍率色収差などの収差による横（像面内）方向ずれ（図１４）がある。勿論光学系の他の収差は通常の顕微鏡用と同様に良好に補正されているとしている。
【００６９】
図１３は、テレセントリックで無い事による試料凹凸やピンボケに依存する本実施形態に係る機能出射画素の像の機能受光画素位置からのずれを示す。対物レンズ２４や第２対物レンズ３１がテレセントリックでない場合、画面周辺部でピントずれや試料凹凸に伴い、機能出射画素の像が対応する機能受光画素からずれて受光強度が低下する、強度低下を前述の許容規準以内にする為には焦点深度内でのピンボケ像位置の最大ずれ量を画素寸法（エアリーディスク径）の１／５以下に抑える必要がある。即ち、テレセントリック性を示す最大主光線傾斜角ω’はＤＯＦ’ｘｔａｎω’〜λω’／（２ＮＡ’^２）＜２δ／５＝１．２２λ／（５ＮＡ’）より、次式になる。
【００７０】
第２対物レンズの最大主光線傾斜角ω’＜２・ＮＡ’／５
対物レンズは反射光利用のため半分となり、次式になる。
【００７１】
対物レンズの最大主光線傾斜角ω＜ＮＡ／５
どちらも開口数が小さい程よりテレセントリック性が必要となる。例えば、通常像側開口数が最小の１００ｘ／０．９対物レンズを使用した場合、対物レンズは主光線傾斜角ω＜０．９／５＝０．１８ｒａｄ〜１０°（非テレセントリック）。像側開口数が０．００９なので、第２対物レンズの主光線傾斜角ω’＜０．００９ｘ２／５＝３．６ｍｒａｄ〜１２’（テレセントリック）となる。この条件を満たすように、対物レンズ２４と第２対物レンズ３１をテレセントリックにする。実際にはずれ要因は以下の面内のずれなどもあるため、テレセントリック性単独のずれはより少なくする事が望ましい。
【００７２】
図１４は、像面内での像位置のずれ即ち２次元シャタアレイ２１と２次元検出器３２との間の、倍率誤差、歪曲収差、倍率色収差及び配置誤差等による、機能出射画素像のずれを示している。前述の様に８０％以上の受光強度を得るための条件は、２次元シャタアレイ２１の機能出射画素の光学系による像と２次元検出器３２の機能受光画素の最大ずれ量は画素寸法の１／５以下で、次式になる。
最大歪曲収差＜画素寸法／（５ｘ最大画面半径）
最大倍率色収差＜画素寸法／５
これらの条件を達成する様にコリメータレンズ２２、対物レンズ２４と第２対物レンズ３１などの光学系の倍率色収差と歪曲収差を補正する。例えば、全有効画面が１０００ｘ１０００画素の場合、最大歪曲収差＜１／｛５ｘ（√２ｘ１０００／２）｝〜０．０３％程度以下となる。これは、従来の顕微鏡対物レンズの１／１０程度の歪曲収差量である。実際には、ずれの要因が倍率誤差、歪曲収差、倍率色収差及び配置誤差など複数あるため、個々の単独要因によるずれは画素寸法／５より少なくする事が望まれる。尚、もちろん単色使用の場合には倍率色収差の条件は不要になる。
【００７３】
図１５は、短波長光源１１、励起フィルタ２６、ダイクロイックミラー２７、吸収フィルタ３３で構成される共焦点蛍光顕微鏡の例である。光源１１にレーザなどの単色光使用の場合励起フィルタ２６は不用となる。ダイクロイックミラー２７の代りに４５度のノッチフィルタを使用しても良い。
【００７４】
図１６は、偏光子２８、ノマルスキープリズム２９、波長板３４、検光子３５とからなる共焦点偏光顕微鏡と微分干渉顕微鏡の例である。ノマルスキープリズム２９は微分干渉顕微鏡に使用する。偏光子２８、検光子３５の代りに偏光ビームスプリッタを使用しても良い。
【００７５】
なお、２次元検出器３２と２次元シャッタアレイ２１との同期走査を行う際に、例えば、２次元検出器３２の同期走査を２次元シャッタアレイ２１より遅らせることにより、動態計測、蛍光寿命、蛍光偏光解消の計測などが可能となる。
【００７６】
尚、本発明は、上記の実施形態に限定されるものではない。その要旨の範囲内で数々の変形が可能である。例えば、上記の一実施形態において、新たな構成要素を加えることで、カラー、位相差等の各種方式の共焦点顕微鏡も実現が容易となる。
【００７７】
【発明の効果】
以上のように、本発明によれば、小型化された共焦点顕微鏡を提供できる。
【図面の簡単な説明】
【図１】本発明の実施形態に係る共焦点顕微鏡の構成図である。
【図２】本発明の実施形態に係る共焦点顕微鏡の多点走査（２次元シャッタアレイ２１と２次元検出器３２との同期走査）における機能画素パターンを示す図である。
【図３】本発明の実施形態に係る共焦点顕微鏡における機能画素パターン（正方配置）と多点走査の走査方式を説明するための図である。
【図４】本発明の実施形態に係る共焦点顕微鏡における機能画素パターン（六方配置）と多点走査の走査方式を説明するための図である。
【図５】本発明の実施形態に係る共焦点顕微鏡における機能画素パターン（矩形配置）と多点走査の走査方式を説明するための図である。
【図６】本発明の実施形態に係る共焦点顕微鏡における機能画素パターン（交互配置）と多点走査の走査方式を説明するための図である。
【図７】本発明の実施形態に係る共焦点顕微鏡における全有効画面の縦横比例の交互配置を説明するための図である。
【図８】本発明の実施形態に係る共焦点顕微鏡における２ｘ２画素構成の機能画素で全有効画面の縦横比例の矩形配置を説明するための図である。
【図９】本発明の実施形態に係る共焦点顕微鏡における３ｘ３画素構成の機能画素で間隔５の正方配置を説明するための図である。
【図１０】本発明の実施形態に係る共焦点顕微鏡における５ｘ５画素構成の機能画素で間隔６の六方配置を説明するための図である。
【図１１】本発明の実施形態に係る共焦点顕微鏡におけるその他の機能画素の配置例を示す図である。
【図１２】本発明の実施形態に係る共焦点顕微鏡における機能画素間隔とピンボケ像によるクロストークを説明するための図である。
【図１３】本発明の実施形態に係る共焦点顕微鏡における共焦点光学系のテレセントリック性と像位置のずれとの関係を説明する図である。
【図１４】本発明の実施形態に係る共焦点顕微鏡における２次元シャッタアレイと２次元検出器との間の像のずれを説明する図である。
【図１５】本発明の別の実施形態に係る共焦点顕微鏡の構成図である。
【図１６】本発明の別の実施形態に係る共焦点顕微鏡の構成図である。
【符号の説明】
１…照明系、２…送光系、３…受光系、４…制御系、１１…光源、１２…淡色フィルタ、１３…集光レンズ、２１…２次元シャッタアレイ、２２…コリメータレンズ、２３…ビームスプリッタ、２４…対物レンズ、２５…試料、２６…励起フィルタ、２７…ダイクロイックミラー、２８…偏光子、２９…ノマルスキープリズム、３１…第２対物レンズ、３２…２次元検出器、３３…吸収フィルタ、３４…波長板、３５…検光子、４０…制御装置、２１１…機能出射画素（シャッタ）、２１２…遊休出射画素（シャッタ）、３２１…機能受光画素（センサ）、３２２…遊休受光画素（センサ）。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a confocal microscope.
[0002]
[Prior art]
The confocal microscope improves in-plane resolution and on-axis resolution (depth of focus) by 20 to 30% as compared with a conventional so-called imaging microscope, and has no out-of-focus image noise of the sample or flare of the optical system. Since a three-dimensional image can be easily obtained with high contrast, it is widely used in many fields (in particular, industrial fields and biological / fluorescent observation).
[0003]
As a typical example of the conventional confocal microscope, there is a method of mechanically scanning a light beam with a rotating Nipkow disk and receiving the light beam with a two-dimensional detector (for example, see Patent Document 1).
[0004]
In this method, a plurality of light transmitting pinholes formed on a rotating Nipkow disk are spirally arranged at predetermined intervals. The light emitted from the light source is converted into a parallel light by an optical system including a reflecting mirror and a lens, and is irradiated on one of the pinholes. A part of the irradiation light is transmitted through the pinhole, and then irradiated on the sample through the objective lens. A pinhole traverses the beam as the Nipkow disk rotates, and the light transmitted through the pinhole scans an arc-shaped region of the sample. Further, as the Nipkow disk rotates, light transmitted through the spirally arranged pinholes sequentially scans another region of the sample in an arc shape. As a result, the sample can be two-dimensionally scanned with the light transmitted through the pinhole. The reflected light from the sample is detected by a two-dimensional detector or the like.
[0005]
[Patent Document 1]
JP-A-9-329748 "Confocal microscope"
[0006]
[Problems to be solved by the invention]
However, the conventional confocal microscope as described above requires a mechanical light beam scanning system as compared with an imaging microscope, and thus has a problem in that the apparatus becomes large.
[0007]
The present invention has been made in view of such a conventional problem, and has as its object to provide a confocal microscope that can be downsized.
[0008]
[Means for Solving the Problems]
In order to solve the above problems, the invention according to claim 1 is:
A plurality of emission pixels capable of emitting light, emission means arranged two-dimensionally,
Light receiving means having a plurality of light receiving pixels corresponding to the plurality of emission pixels of the emission means,
A confocal optical system that irradiates a sample with light emitted from the emission unit and condenses return light from the sample surface on the light receiving unit,
One of the plurality of emission pixels or a plurality of adjacent pixels is set as a function emission pixel, and at least two of the plurality of function emission pixels are set at non-adjacent positions. A light receiving pixel corresponding to the function emitting pixel is set as a function light receiving pixel, and a process of emitting light from the function emitting pixel and receiving the return light at the function light receiving pixel is performed by a different one of the plurality of emission pixels. Control means for obtaining an image of the sample by performing a plurality of times on emission pixels or functional emission pixels set in different combinations of emission pixels.
[0009]
The invention according to claim 2 is
The confocal microscope according to claim 1, wherein
An interval between the at least two functional emission pixels and an interval between at least two functional light receiving pixels corresponding to the at least two functional emission pixels are more than 10 times and less than 100 times the diameter of each Airy disk. .
[0010]
The invention according to claim 3 is:
The confocal microscope according to claim 1, wherein
The control means sets the interval between the at least two functional emission pixels and the interval between at least two functional light receiving pixels corresponding to the at least two functional emission pixels to be larger as the step of the unevenness on the sample surface is larger. Features.
[0011]
The invention according to claim 4 is
The confocal microscope according to claim 1, wherein
The control means sets an interval between the at least two functional emission pixels and an interval in pixel units of at least two functional light receiving pixels corresponding to the at least two functional emission pixels,
(Step of unevenness of sample) / (2.44 × depth of focus)
It is characterized in that it is set larger.
[0012]
The invention according to claim 5 is
The confocal microscope according to claim 1, wherein
The control means changes the number of pixels constituting the functional emission pixels and the functional light receiving pixels according to the Airy disk diameter depending on the pupil diameter of the objective lens used.
The invention according to claim 6 is
The confocal microscope according to claim 1, wherein
The control unit may perform a process of emitting light from the functional emission pixel and causing the functional light receiving pixel to receive the return light for a plurality of functional emission pixels set by different combinations of emission pixels among the plurality of emission pixels. When performing the number of times, the distance between the centers of the functional emission pixels set a plurality of times is set to one emission pixel or less than the number of vertical or horizontal emission pixels constituting the functional emission pixel. It is characterized by the following.
[0013]
The invention according to claim 7 is
The confocal microscope according to claim 1, wherein
The control means may arrange the plurality of functional emission pixels and the light receiving function pixels in a square arrangement, a hexagonal arrangement, a rectangular arrangement having a ratio equal to the aspect ratio of the entire effective screen, or an alternating arrangement having a ratio equal to the aspect ratio of the entire effective screen. It is characterized by the following.
[0014]
The invention according to claim 8 is
The confocal microscope according to claim 1, wherein
The control unit performs a process of emitting light from the functional emission pixel and causing the functional light receiving pixel to receive the return light a plurality of times for functional emission pixels set by different emission pixels among the plurality of emission pixels. Thus, when obtaining the image of the sample, the image of the sample is formed by thinning scanning and interpolation.
[0015]
The invention according to claim 9 is
The confocal microscope according to claim 1, wherein
The confocal optical system,
It is characterized by having a collimator lens, a first objective lens and a second objective lens whose maximum distortion and maximum chromatic aberration of magnification are corrected to 1/5 or less of the light receiving pixel size.
[0016]
The invention according to claim 10 is
The confocal microscope according to claim 9, wherein
The first objective lens and the second objective lens are characterized in that the maximum chief ray inclination angle is corrected to an angle such that the amount of deviation within the depth of focus is within 1/5 of the size of the light receiving pixel. I do.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of the present invention will be described.
[0018]
FIG. 1 is a schematic configuration diagram of a confocal microscope according to one embodiment of the present invention.
[0019]
The confocal microscope includes an illumination system 1 having a light source 11, a monochromatic filter 12, and a condenser lens 13, a two-dimensional shutter array 21, a collimator lens 22, a beam splitter 23, and an infinity objective lens 24 (including a finite objective lens). (Which can be configured), a light receiving system 3 having a second objective lens 31 and a two-dimensional detector 32, and a control device 40 for controlling the two-dimensional shutter array 21 and the two-dimensional detector 32. A control system 4.
[0020]
Here, the two-dimensional shutter array 21 is located at the focal position of the collimator lens 22. In the present embodiment, the two-dimensional shutter array 21 is formed of a two-dimensional digital micromirror device (DMD), and has a plurality of pixels that reflect and emit light from the illumination system 1. The pixels of the two-dimensional shutter array 21 are called emission pixels. When one or a plurality of continuous emission pixels emit light to the collimator lens 22, the emission pixel is called a function emission pixel 211, and when the light is not emitted, the idle emission pixel 212 is used. I will call it.
[0021]
The two-dimensional shutter array 21 can be replaced with a high-speed two-dimensional liquid crystal shutter (transmission system) or the like.
[0022]
As the light source 11, a laser, a normal high-luminance light source (for example, a mercury lamp, a metal halide), or the like can be used. When a two-dimensional digital micromirror device (DMD) is used for the two-dimensional shutter array 21, an array of lasers or LEDs that emit pulses synchronized with the DMD may be used for the light source 11. In this case, a pulsed optical signal may be intermittently transmitted in accordance with the change of the mirror pattern of the DMD (irradiates the functional pixel 211), and the light use efficiency is improved as compared with other light sources that continue to irradiate. .
[0023]
When a two-dimensional laser diode (LD) array is used, the two-dimensional shutter array 21 and the illumination system 1 can be omitted, resulting in a smaller, lighter, and more efficient device.
[0024]
The two-dimensional detector 32 is located at the focal position of the second objective lens 31. The two-dimensional detector 32 has a plurality of light receiving pixels corresponding to a plurality of emission pixels in the two-dimensional shutter array 21. One or more continuous pixels in which the light receiving function of the light receiving pixel is working are called functional light receiving pixels 321, and pixels in which the light receiving function is not working are called idle light receiving pixels 322.
[0025]
The optical system including the collimator lens 22, the beam splitter 23, the objective lens 24, and the second objective lens 31 is connected to the functional output pixel 211 of the two-dimensional shutter array 21, the sample 25, and the functional light receiving pixel 321 on the two-dimensional detector 32. On the other hand, it is a confocal optical system. That is, in the confocal optical system, the illumination system includes the collimator lens 22 and the imaging system includes the second objective lens 31. The beam splitter 23 and the objective lens 24 are shared by the illumination system and the imaging system. Place the sample at the focal position.
[0026]
Next, the operation of the confocal microscope having the above configuration will be described.
[0027]
The light emitted from the light source 11 passes through the monochromatic filter 12 and is condensed on the two-dimensional shutter array 21 by the condenser lens 13. The light reflected by the function emission pixel 211 on the two-dimensional shutter array 21 is converted into a parallel light by the collimator lens 22, passes through the beam splitter 23, and reaches the objective lens 24. The light incident on the objective lens 24 irradiates the sample 25.
[0028]
The reflected light from the sample 25 is converted into a parallel light beam by the objective lens 24, reflected by the beam splitter 23, and reaches the second objective lens 31. The reflected light from the sample 25 is detected by the functional light receiving pixel 321 of the two-dimensional detector 32 as a confocal image corresponding to the functional emission pixel 211. In the confocal microscope of the present embodiment, the functional emission pixel functions as an emission confocal pinhole, and the functional light receiving pixel functions as a light receiving confocal pinhole. Hereinafter, in this specification, when simply referred to as “functional pixel”, it refers to both the functional emission pixel and the functional light receiving pixel.
[0029]
In order to obtain an image of the entire sample 25, the following operation is performed on the functional pixels 211 and 321 located at conjugate positions through control software on which the control device 40 is mounted.
[0030]
The control device 40 causes a plurality of pixels among the emission pixels in the two-dimensional shutter array 21 to function, and transmits light to the light transmission system 2. On the other hand, the control device 40 selects, from the light receiving pixels in the two-dimensional detector 32, light receiving pixels corresponding to a plurality of pixels in the transmitted two-dimensional shutter array 21. Then, the selected plurality of light receiving pixels are made to function to receive light, and images of the sample 25 corresponding to the plurality of emission pixels are obtained. Therefore, confocal image information on a plurality of samples 25 can be obtained by one scan. This processing (referred to as simultaneous multipoint processing) is performed a plurality of times until all pixels are covered, and an image of the entire sample 25 is obtained.
[0031]
In order to obtain a confocal image corresponding to all emission pixels in the above configuration, first, the use area and the pixel shape of the two-dimensional shutter array 21 and the two-dimensional detector 32 are similar to each other, and the number of pixels in the use area is the same. It is preferred that Further, the focal length of the collimator lens 22, ie, the magnification, and the focal length of the second objective lens 31 are set such that the light emitted from each pixel of the two-dimensional shutter array 21 enters each corresponding light-receiving pixel of the two-dimensional detector 32. That is, the magnification is determined.
[0032]
As is well known, the optimal functional pixel size of the two-dimensional shutter array 21 (the diameter of the circle inscribed in the pixel, that is, the optimal confocal pinhole diameter) is that the image has high resolution, high sensitivity, and high efficiency. In addition, it is desirable that the diameter is approximately the same as the Airy disk diameter. This is because the resolution decreases when the pixel size exceeds the Airy disk diameter, and the sensitivity and efficiency decrease when the pixel size is smaller than the Airy disk diameter. Accordingly, the focal length, that is, the magnification of the collimator lens 22 is determined from the object-side numerical aperture (NA) of the objective lens 24 and the focal length so that the optimum pixel size is obtained.
[0033]
With the above-described configuration, a large and expensive mechanical beam scanning mechanism is eliminated, and the entire apparatus is manufactured by using an inexpensive two-dimensional shutter array (for example, a DMD) 21 and a two-dimensional detector 32 by mass production effects. Small, lightweight and inexpensive. Further, high-speed scanning / imaging at a TV (video) rate (30 images per second) or more is realized by imaging using the multipoint scanning method. Since a mechanical beam scanning mechanism is not required, there is no noise, uneven scanning light intensity, or image distortion due to linear or rotary scanning, and a low-magnification, high-numerical-aperture objective lens can be used. Further, since the optimum confocal pinhole diameter can be quickly changed by the control software provided in the control device 40, the pupil diameter (= 2 × focal length × numerical aperture) is different, so that the objective lenses of various types having different Airy disk diameters are different. It has the effect that it can be used and the range of use is greatly expanded.
[0034]
FIG. 2 is a diagram in which M functional pixel patterns (1 to M) used in synchronous multipoint scanning between the two-dimensional shutter array 21 and the two-dimensional detector 32 are arranged along the time axis. In the following, the function emitting pixel and the function light receiving pixel always function synchronously with the same shape and arrangement, and are therefore collectively referred to as function pixels.
[0035]
In the multi-point scanning, there are (the total number of pixels / the number of functional pixels) functional pixel patterns (M). An entire image is obtained by repeating multipoint scanning by the number of functional pixel patterns. For example, if the total number of pixels is 1 million, and if the number of functional pixels is 10,000, one image can be obtained by scanning 100 times. In the equidistant array of the functional pixels of the multi-point scanning, the functional pixel interval in functional pixel units is defined as Px and Py in the horizontal direction and the vertical direction is defined as Px / Py = functional pixel interval / functional pixel (natural number), and the displacement amount of the column is D = column It can be displayed as a shift / functional pixel (zero, natural number) (see FIG. 7). The number of functional pixel patterns constituting one image is Px · Py, and the number of functional pixels is the total number of pixels / the number of functional pixel patterns (M). Since the entire screen is a repetition of the scan range of Px and Py including one functional pixel, this is called a unit scan range. The column shift D is also the horizontal shift of the column unit scanning range. (See FIGS. 3 to 6) Of various functional pixel arrays, Px = Py = P, D = 0 is called a square arrangement for convenience (see FIG. 3), Px is an even number, and Py = Px · sin (π). / 3) to the nearest natural number, D = Px / 2, is called hexagonal arrangement (see FIG. 4). Further, assuming that the number of square pixels in the vertical and horizontal directions of the entire effective screen is Ty and Tx, the aspect ratio Ty: Tx = Py: Px, and D = 0 is called a rectangular arrangement proportional to the entire effective screen (see FIG. 5). : Tx = Py: Px, D = Px / 2-The most recent natural number is referred to as an alternate arrangement proportional to the entire effective screen (see FIG. 6).
[0036]
FIGS. 3 to 6 show examples of the entire effective screen shape, the spacing and arrangement of the functional pixels, and a scanning example of the confocal microscope according to the present embodiment. In the following example, the number of pixels will be described for simplicity. However, in an actual device configuration, the number of pixels of the two-dimensional shutter array and the two-dimensional detector is usually about Tx to Ty to 1000, and the functional pixel intervals are also Px and Py. = About 10 to 100.
[0037]
FIG. 3 shows a specific example of a functional pixel (confocal pinhole) pattern and a scanning method over time. In this figure, the number of functional pixels is 4 in an example where the total effective screen Tx · Ty = 12 × 12 pixels, the square arrangement Px = Py, the functional pixel interval P = 6, and D = 0. One image can be obtained by scanning PxP = 36 patterns. In the multi-point scanning, the position of the functional pixel is shifted with time while the relative functional pixel arrangement is maintained. A portion shown by a thin solid line is a unit scanning range including one functional pixel, and one image is formed by scanning all the functional pixels once in each unit scanning range. Reference numeral 3A denotes sequential scanning in which the pixel is shifted by one pixel, 3B denotes interlaced scanning in which the pixel is shifted by two pixels, and interlaced scanning in which the pixel is scanned again by shifting one pixel. Indicates thinning-out scanning in which scanning is performed while obliquely shifting one pixel and an insufficient image is interpolated. In particular, thinning-out scanning is effective for high-speed imaging of dynamics, imaging of three-dimensional images, and the like. In other words, if thinning-out imaging of one to several pixels is performed first and then the images are completed collectively by interpolation processing, an imaging speed two to several times faster than sequential scanning can be realized.
[0038]
FIG. 4 shows an example in which the entire screen Tx · Ty = 12 × 12 pixels, hexagonal arrangement, functional pixel interval P = 6, and D = 3. Since the area of the entire effective screen is not a multiple of the area of the unit scanning range indicated by a thin solid line, the number of functional pixels changes from 4 to 6 during scanning. One image is composed of Px · Py = PxP · sin (π / 3) to 6 × 5 = 30 functional pixel patterns. 4A to 4D show sequential scanning, one-pixel skipping scanning, and thinning scanning. The hexagonal arrangement has better image symmetry than the square arrangement at the same interval (described later), and the number of functional pixel patterns is small, and high-speed imaging is possible.
[0039]
FIG. 5 shows an example of a rectangular arrangement in which the entire effective screen Ty · Tx = 9 × 12 pixels, the aspect ratio Py: Px = 3: 4, and D = 0 in proportion to the entire screen, and the number of functional pixels is 9 × 12/3 × 4 = 9. One image is composed of Px · Py = 12 functional pixel patterns. Reference numerals 5A to 5D denote sequential scanning, one-pixel skipping scanning, and thinning scanning. Although the rectangular arrangement in proportion to the entire effective screen facilitates image acquisition and processing, the symmetry of the image is low (described later).
[0040]
FIG. 6 shows an example of an alternate arrangement of the entire effective screen Ty · Tx = 9 × 12 pixels, the aspect ratio Py: Px = 3: 4, and D = Px / 2 = 2 in proportion to the entire screen. The functional pixels are 9 × 12/3 × 4 = 9. is there. One image is composed of a scanning pattern of Px · Py = 12. 6A to 6D show sequential scanning, one-pixel skipping scanning, and thinning scanning. Since the arrangement is proportional to the entire screen, image acquisition and processing are easy, and the symmetry of the image is higher than that of the rectangular arrangement at the same interval (described later).
[0041]
7 to 11 show dimensions of output pixels, intervals between functional output pixels, arrangement examples, and features of the confocal microscope according to the present embodiment.
[0042]
In the multi-point scanning, a large number of functional pixels are present at the same time, and an out-of-focus image due to the unevenness of the sample becomes noise (described later). If the distance between the functional pixels is different depending on the direction, the noise of the image, that is, the contrast, becomes directional. The symmetry of this image can be roughly evaluated by the following equation S.
[0043]
S = minimum / maximum contrast ratio = maximum / minimum noise ratio = (minimum interval / maximum interval)²S = P for square arrangement²/ (P²+ P²) = 0.5, in the hexagonal arrangement, the spacing in all directions is constant and S = ｛(Px / 2)²+ (Pxsin (π / 3))²) / Px²= 1 Therefore, the hexagonal arrangement is an arrangement with good symmetry without a difference in contrast depending on the direction. The effect of image symmetry is small when the functional pixel interval P is small, and is large when observing a low-contrast sample, but can be neglected in the opposite case.
[0044]
FIG. 7 shows an example of an alternate arrangement of all effective screens Ty · Tx = 27 × 48 pixels, similar to all effective screens of the HDTV system, and a practical effect can be expected. The aspect ratio is Ty: Tx = Py: Px = 9: 16, D = Px / 2 = 8, and the number of functional pixels is 27 × 48/9 × 16 = 9. One image is composed of a functional pixel pattern of Px · Py = 144. The two arrows and the formulas between the functional pixels in the figure indicate the shortest and longest functional pixel (confocal pinhole) intervals, and the symmetry of the image is S = (D²+ Py²) / Px²~ 0.57.
[0045]
FIG. 8 shows an example of a 3: 4 rectangular arrangement (Py = 6, Px = 8, D = 0), which is a functional pixel composed of 2 × 2 pixels and similar to all effective screens of TV and video (Ty · Tx = 36 × 48). The above effect can be expected. The functional pixel (diameter of confocal pinhole) has a 2 × 2 pixel configuration, and is suitable for use of an objective lens having a middle and large pupil diameter. Because of the rectangular arrangement, the symmetry of the image is S = Py²/ (Px²+ Py²) = 0.36, which is low.
[0046]
In the multi-point scanning, when one functional pixel is composed of one pixel, scanning is performed at a pitch of one functional pixel (one pixel). However, when one functional pixel is composed of a plurality of pixels such as a 2 × 2 pixel configuration, scanning is performed. The pitch can be variably controlled from one pixel to one functional pixel (two pixels). Resolution is improved by scanning at a pitch finer than one functional pixel pitch, but the number of scanning patterns is larger than Px · Py, and the imaging time is longer. Normally, in order to achieve a good resolution, the function pixels corresponding to the minimum Airy disk diameter are configured in a 2 × 2 pixel configuration, and scanning is performed at one pixel pitch. In this case, the resolution is about twice as large as the scanning at the pitch of one functional pixel (two pixels), but the number of scanning patterns is four times and the imaging time is also four times. The scanning pitch of the functional pixels can be easily selected and controlled by software depending on the purpose and the sample.
[0047]
In general, the optimal diameter of the confocal pinhole (functional pixel) is about the Airy disk diameter (as described above). If the wavelength is λ, the refractive index is n, the aperture angle of the objective lens is θ, the numerical aperture is NA = nsinθ, and the magnification is β, the image-side numerical aperture is NA ′ = n′sinθ ′ = NA / β, and the image-side numerical aperture is (Functional Pixel Side) The Airy disk diameter is 2δ = 1.22λ / NA ′ = 1.22λxβ / NA. If an objective lens having a different magnification or type is used, the image-side numerical aperture (pupil diameter) greatly changes, and thus the Airy disk diameter also changes. When the magnification and numerical aperture of the objective lens are expressed as magnification x / numerical aperture, the actual objective lens is 5 × / 0.25, 10 × / 0.3, 100 × / 0.9, etc., and NA ′ = 0.05. , 0.03, and 0.009, and the Airy disk diameters are also significantly different from about 1, 3.3, and 5.6, respectively, on a five-fold basis. Therefore, when using various types of objective lenses, it is necessary to increase or decrease the number of pixels constituting one functional pixel in order to adjust the optimal confocal pinhole diameter to the Airy disk diameter. Specifically, when the functional pixels of the objective lens of 5x / 0.25 are configured as 2x2 pixels (scanning pitch is 1 pixel), the functional pixels of 10x / 0.3 are configured as 3x3 pixels, and the function of 100x / 0.9 is used. Pixels are arranged in an 11 × 11 pixel configuration.
[0048]
In particular, with a low-magnification, medium-numerical-aperture objective lens, which has been difficult to use simultaneously with an objective lens having a small image-side numerical aperture, a wide range of a sample can be observed and measured with high resolution, and the efficiency is improved. That is, when comparing objective lenses having the same numerical aperture and differing in magnification by a factor of two, an objective lens having a magnification of half can observe and measure a four-fold wider range, and the efficiency is increased by a factor of four.
[0049]
In the three-dimensional shape inspection of a micro component such as a printed wiring board, a diffractive optical element, and a MEMS (Micro Electro Mechanical System), the design data is manually compared with the result of the three-dimensional shape inspection using a confocal microscope, and the quality of the micro component is determined. Is determined. In a general inspection process, since these parts have various shapes and sizes, it is necessary to switch to and use various types of objective lenses having different image-side numerical apertures according to the types of the parts. However, in the conventional method using a Nipkow disk, it is not easy to change the diameter of the pinhole formed in the Nipkow disk. Therefore, with a conventional confocal microscope, only a limited range of objective lenses can be used effectively.
[0050]
In the confocal microscope of the present embodiment, the dimensions of the functional pixels (optimal confocal pinhole diameters) can be changed and selected only by increasing or decreasing the number of constituent pixels of the functional pixels under the control of software of the control device 40. Quick response to use of objective lens. In particular, by combining a low-magnification medium-numerical-aperture objective lens, the shape of various minute components can be inspected more quickly and efficiently than before, while controlling the contrast (described later). Further, the entire comparison inspection process can be performed quickly and efficiently by automating the comparison between the design data converted into the three-dimensional CAD data and the three-dimensional shape measurement result by the confocal microscope of the present invention.
[0051]
FIG. 9 shows an example of a 5 × 5 pixel configuration, a total effective screen Tx · Ty = 45 × 45 pixels, and a square arrangement (Px = Py). From the functional pixel interval P = 5 and D = 0, one image is composed of a PxP = 25 functional pixel pattern. Image symmetry S = P²/ (P²+ P²) = 0.5 This is an example suitable for an objective lens having a small pupil diameter such as an objective lens of 100x / 0.9.
[0052]
FIG. 10 shows an example of a hexagonal arrangement with a 3 × 3 pixel configuration and all effective screens Tx = Ty = 45 × 45 pixels. From the functional pixel interval Px = 6, Py = Px · sin (π / 3) to 5 (a recent natural number), and D = Px / 2 = 3, one image is composed of Px · Py = 30 functional pixel patterns. Image symmetry S = (D²+ Py²) / Px²= 0.94, which is lower than the ideal hexagonal arrangement of 1, but there is no practical problem. This is an example suitable for an objective lens having a relatively small pupil diameter such as a 50x / 0.8 objective lens.
[0053]
FIG. 11 shows an example of another functional pixel array in the case of a configuration of all effective screens Tx and Ty = 15 × 15 pixels. The upper row is a point arrangement, that is, a point-scanning confocal microscope. Can be The functional pixel interval is Px = Py = P = Tx = Ty = 15. The middle stage has a horizontal multi-line arrangement, which enables higher-speed imaging than a line-scanning confocal microscope. The functional pixel interval is Px = 1, Py = 5, and the image symmetry S = Px²/ Py²<0.04 is the lowest. The lower stage shows the case where all are function pixels, that is, the case of an imaging microscope. The highest imaging speed can be realized, but the contrast of the image is the lowest. The functional pixel interval is Px = Py = P = 1. In the present embodiment, the arrangement of the functional pixels is not changed by hardware, and the control software provided in the control device 40 uses point-and-line, multi-point scanning, and image-forming (non-light scanning) or the like depending on the purpose and the sample. The scanning method can be selected and controlled.
[0054]
In multi-point scanning, since a large number of functional pixels are present at the same time, the light receiving noise increases or decreases depending on the intervals and arrangement of the functional pixels. Although the symmetry of the image by the arrangement has been described above, the relationship between the functional pixel interval and the noise and the practically optimal functional pixel interval range will be described below.
[0055]
FIG. 12 is a cross-sectional view of the two-dimensional detector at a certain time, and illustrates a relationship between a functional pixel interval and a light amount crosstalk due to an out-of-focus image. In the case of a thick sample or a sample having large irregularities, light amount crosstalk leaking to an adjacent functional pixel due to an out-of-focus image appears. When the out-of-focus image reaches an adjacent functional pixel, light amount noise occurs and the contrast is reduced. Leaks to adjacent idle pixels are not detected and can be ignored. In the following, for simplicity, the case where the functional pixels are arranged in a square (Px = Py = P) will be considered based on the imaging type (P = 1). The depth of focus on the image side is DOF ′ = ± λ / (2NA ′)²), The actual functional pixel interval is Pr = Px1.22λ / NA ′, n ′ = 1, tan θ ′ to sin θ ′ = NA ′, and from FIG. 12, the blur amount (distance on the optical axis) that does not leak to the adjacent functional pixels is η_P'= Px1.22xλ / NA'²= Px2.44xDOF '. The corresponding sample unevenness (sample unevenness amount without defocus noise) is η_PThen η_P= Η_P’/ Β², NA = β · NA ′, the following equation that is proportional to the functional pixel interval P holds.
[0056]
η_P= Px1.22xλ / NA²= Px2.44xDOF to Px2xDOF
For example, in the imaging type (P = 1), η₁= 1.22 × λ / NA²= 2.44 × DOF, η for point-scan confocal (P = Tx = Ty)_Tx= Tx · 2.44xDOF, but since there is only one functional pixel, there is no leakage light from the adjacent pixel and it is always η regardless of unevenness._Tx= ∞, and there is no out-of-focus noise. Also, let η be the roughness of the sample, and η> η_PIn the case of, the out-of-focus noise (light amount) is²Is inversely proportional to Further, by controlling the functional pixel interval P as follows, it is possible to perform imaging without out-of-focus noise.
[0057]
P> η / (2.44 × DOF) = η ′ / (2.44 × DOF ′)
In the case of a sample whose specific concavities and convexities are unknown, it is efficient to perform high-speed imaging by first reducing the functional pixel interval P and then increase the functional pixel interval P to perform imaging.
[0058]
The following relationship exists between the functional pixel interval P and the image contrast C. I₀Is the irradiation light amount, the noise due to the out-of-focus image is N_F∝I₀/ P²Η> η_P, N_F= 0; η <η_PFlare noise due to reflection of the optical system is N_OＩ I₀/ P², The noise of the combined image is
[0059]
N = N_F + N_OＩ I₀/ P²
The contrast C of the image is determined by the image intensity I∝I₀Then, in the case of N / I << 1, the following equation is obtained.
[0060]
C = (IN) / (I + N) ∝1-2N / I∝1-2 / P²
That is, as the functional pixel interval P is larger (the number of functional pixels is smaller), the noise and flare of the multi-point scanning confocal optical system are reduced, and the contrast C is increased.
[0061]
On the other hand, the imaging speed V is the number of patterns P²Is inversely proportional to
[0062]
V∝I₀/ P²
In the case of the multi-point scanning method, the light utilization efficiency E is given by the following equation because the entire two-dimensional shutter array is uniformly irradiated with light and only the function emission pixels use light.
[0063]
E∝1 / P²
In the imaging type (P = 1), the contrast C is the minimum, and the imaging speed V and the light use efficiency E are the maximum. Although the contrast increases as P increases, the imaging speed and light use efficiency decrease by the square. From the above, it is preferable to balance sample unevenness without defocus noise, image contrast and imaging speed, and light use efficiency.
[0064]
As a specific example, when P = 10 (10 × 10 = one per 100 pixels is used as a functional pixel), a 20 × / 0.75 objective lens, and a center wavelength λ of visible light = 550 nm, the noise-free unevenness is calculated from the above equation. η to 0.012 mm, and the depth of focus (vertical resolution of unevenness) DOF = ± λ / (2NA²) = 0.49 μm and about 27 times spatial resolution δ = 0.61λ / NA = 0.45 μm. Even in the case of a sample having irregularities exceeding noise-free irregularities, the out-of-focus noise compared to the imaging type (P = 1) is 1 / P²= 1/100, and the amount of light leaking to the adjacent functional pixels is small and functions sufficiently as a confocal optical system. Further, the imaging time is not long and practical.
[0065]
On the other hand, when P = 100 (one per 100 × 100 = 10000 pixels is used as a functional pixel), when the same 20 × / 0.75 objective lens and wavelength λ = 550 nm are used, the noiseless unevenness is η to 0. .12 mm and 10 times larger than in the case of P = 10. In other words, although the resolution of the image is the same, it is possible to image up to 10 times thicker sample unevenness without noise. Further, even in the case of a sample having irregularities exceeding this thickness, the noise is 1 / 10,000 as compared with the imaging type, and 1/100 as compared with the case where P = 10. Therefore, it is possible to image a very low contrast sample (a sample having a small difference in reflectance and transmittance). Thus, when P = 100, the contrast of the image is greatly improved. However, the imaging time is 100 times longer than in the case of P = 10, and the light use efficiency is also reduced to about 1/100 in comparison with the case of P = 10. For this reason, first, the whole of a two-dimensional or three-dimensional sample is imaged at high speed (P = 10), and then only parts deemed necessary are imaged again with high contrast and high image quality at P = 100 over time. In addition, it is assumed that a high-contrast imaging in which noise is reduced by multiplying the imaging time by P = 100 for a sample that is known as a low-contrast sample in advance. When the functional pixel interval P is more than 100, the imaging becomes higher in contrast, but the imaging speed and the light use efficiency are further reduced (<1/10000), which is not practical. Needless to say, in the present embodiment, the sample having a known unevenness and thickness can be changed to the optimal functional pixel interval P beyond the range of the functional pixel interval P = 10 to 100 as necessary. In any case, the function pixel interval P can be easily changed by the control software provided in the control device 40.
[0066]
As described above, by changing the value of the functional pixel interval P depending on the purpose and the sample, high-speed imaging (P = 10: low-contrast imaging is permitted), standard imaging (P to 30), and high-contrast imaging (P = 100: low-speed imaging) Can be selected). It is also possible to perform high-speed imaging (P = 10) of the whole of a two-dimensional or three-dimensional sample first, and then perform high-contrast / high-quality imaging (P = 100) of only the necessary portions again. Therefore, the practical functional pixel interval is about P = 10 to 100, and it is appropriate to use it properly depending on the purpose and the sample.
[0067]
As described above, in the multi-point scanning confocal microscope according to the present embodiment, the optimum resolution / efficiency and the selection / control of the objective lens to be used are determined by the functional pixel size (the number of pixels constituting one functional pixel) by software. Selection and control of resolution by scanning pitch; control of imaging speed, contrast (noise) and light use efficiency by functional pixel spacing; control of image symmetry by functional pixel arrangement; scanning methods such as sequential and thinning , The imaging speed can be controlled, and these can be quickly, easily changed, selected, and used depending on the purpose and the form of the sample. Specifically, the size of the functional pixels is about the Airy disk diameter, the number of constituent pixels is 2 × 2, the scanning pitch is 1 pixel, and the interval is about 10 to 100 times the Airy disk diameter, depending on the purpose and sample. An image with an optimum imaging speed, light use efficiency, high resolution and high contrast can be selected. In the case of a sample having a known thickness or unevenness, such as a micro component such as a printed wiring board, a diffractive optical element, or a MEMS (Micro Electro Mechanical System), the functional pixel interval is set to P> η / (2.44 × DOF), thereby defocusing. High contrast and efficient imaging with zero image noise. By arranging the functional pixels in a hexagonal arrangement, a square (square) arrangement, a rectangular arrangement, or an alternate arrangement, or by performing the thinning-out scanning / interpolation processing, it is possible to capture an image with good symmetry, and to perform quicker and more efficient imaging. . Further, by controlling the functional pixel interval to be large, it is possible to control and control an imaging speed, and to select and use an arbitrary imaging method from a point scan to a multipoint scan to an imaging microscope. In the above description, the size and the arrangement of the functional pixels have been collectively described. However, when the size and the arrangement of the functional output pixels are changed, it is needless to say that the size and the arrangement of the functional light receiving pixels are also changed accordingly.
[0068]
Next, the requirements for the optical system of the present embodiment will be described with reference to FIGS. In the present embodiment, conditions that the optical image satisfies are added over the entire area of the screen due to multi-point scanning. When the position of the image formed on the two-dimensional detector 32 by the functional emission pixel on the two-dimensional shutter array 21 and the position of the corresponding functional light-receiving pixel on the two-dimensional detector 32 deviate, the amount of received light decreases, and the intensity due to out-of-focus occurs. It becomes indistinguishable from a drop, which affects the three-dimensional distortion of the acquired image and the fidelity of the image intensity. In order to avoid this, it is required that the displacement area be 20% or less in order to secure a light receiving amount of 80% or more as an allowable criterion of the displacement, and the displacement of the image position is reduced to 1/5 or less of the pixel size at the maximum. It needs to be suppressed. The image position shift includes a longitudinal (optical axis) direction shift (FIG. 13) due to the telecentricity of the optical system and a horizontal (in-image plane) direction shift (FIG. 14) due to aberrations such as distortion and chromatic aberration of magnification. Of course, other aberrations of the optical system are satisfactorily corrected as in a normal microscope.
[0069]
FIG. 13 shows a shift of the image of the functional emission pixel from the functional light receiving pixel position according to the present embodiment depending on the sample unevenness and out-of-focus due to being not telecentric. When the objective lens 24 or the second objective lens 31 is not telecentric, the image of the functional emission pixel is shifted from the corresponding functional light receiving pixel due to the focus shift or the sample unevenness at the peripheral portion of the screen, and the light receiving intensity is reduced. In order to satisfy the above-mentioned tolerance, it is necessary to suppress the maximum deviation amount of the out-of-focus image position within the depth of focus to 1/5 or less of the pixel size (airy disk diameter). That is, the maximum chief ray inclination angle ω 'indicating telecentricity is DOF'x tan ω' to λω '/ (2NA'²) <2δ / 5 = 1.22λ / (5NA ′).
[0070]
Maximum principal ray tilt angle ω ′ <2 · NA ′ / 5 of the second objective lens
The objective lens is halved due to the use of reflected light, and is given by the following equation.
[0071]
Maximum chief ray inclination angle ω <NA / 5 of objective lens
In both cases, the smaller the numerical aperture, the more telecentricity is required. For example, when a 100x / 0.9 objective lens having the smallest image-side numerical aperture is used, the objective lens has a chief ray tilt angle ω <0.9 / 5 = 0.18 rad to 10 ° (non-telecentric). Since the image-side numerical aperture is 0.009, the chief ray inclination angle ω ′ of the second objective lens <0.009 × 2/5 = 3.6 mrad to 12 ′ (telecentric). The objective lens 24 and the second objective lens 31 are made telecentric so as to satisfy this condition. Actually, there are the following in-plane deviations as deviation factors, so it is desirable to reduce the deviation of the telecentricity alone.
[0072]
FIG. 14 shows the shift of the image position in the image plane, that is, the shift of the functional emission pixel image between the two-dimensional shutter array 21 and the two-dimensional detector 32 due to a magnification error, distortion, magnification chromatic aberration, and an arrangement error. Is shown. As described above, the condition for obtaining the light receiving intensity of 80% or more is that the maximum shift amount between the image of the functional output pixel of the two-dimensional shutter array 21 by the optical system and the functional light receiving pixel of the two-dimensional detector 32 is 1/1 of the pixel size. Below 5 the following equation is obtained.
Maximum distortion <pixel size / (5x maximum screen radius)
Maximum magnification chromatic aberration <pixel size / 5
The chromatic aberration of magnification and distortion of optical systems such as the collimator lens 22, the objective lens 24, and the second objective lens 31 are corrected so as to achieve these conditions. For example, when the entire effective screen is 1000 × 1000 pixels, the maximum distortion is about 1 / {5 × ({2 × 1000/2)}｝ 0.03% or less. This is a distortion amount of about 1/10 of the conventional microscope objective lens. Actually, there are a plurality of factors such as magnification error, distortion, chromatic aberration of magnification, and arrangement error. Therefore, it is desired that each individual factor be smaller than the pixel size / 5. In the case of using a single color, the condition of the chromatic aberration of magnification is not required.
[0073]
FIG. 15 is an example of a confocal fluorescence microscope including the short wavelength light source 11, the excitation filter 26, the dichroic mirror 27, and the absorption filter 33. When monochromatic light such as a laser is used for the light source 11, the excitation filter 26 becomes unnecessary. A 45 degree notch filter may be used instead of the dichroic mirror 27.
[0074]
FIG. 16 shows an example of a confocal polarizing microscope and a differential interference microscope, each of which includes a polarizer 28, a Nomarski prism 29, a wave plate 34, and an analyzer 35. The Nomarski prism 29 is used for a differential interference microscope. A polarizing beam splitter may be used in place of the polarizer 28 and the analyzer 35.
[0075]
When performing synchronous scanning between the two-dimensional detector 32 and the two-dimensional shutter array 21, for example, by delaying the synchronous scanning of the two-dimensional detector 32 from the two-dimensional shutter array 21, dynamic measurement, fluorescence lifetime, fluorescence Measurement of depolarization becomes possible.
[0076]
Note that the present invention is not limited to the above embodiment. Many modifications are possible within the scope of the gist. For example, in the above embodiment, by adding a new component, it is easy to realize confocal microscopes of various types such as color and phase difference.
[0077]
【The invention's effect】
As described above, according to the present invention, a miniaturized confocal microscope can be provided.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a confocal microscope according to an embodiment of the present invention.
FIG. 2 is a diagram showing a functional pixel pattern in multipoint scanning (synchronous scanning between a two-dimensional shutter array 21 and a two-dimensional detector 32) of the confocal microscope according to the embodiment of the present invention.
FIG. 3 is a diagram for explaining a functional pixel pattern (square arrangement) and a scanning method of multipoint scanning in the confocal microscope according to the embodiment of the present invention.
FIG. 4 is a diagram for explaining a functional pixel pattern (hexagonal arrangement) and a scanning method of multipoint scanning in the confocal microscope according to the embodiment of the present invention.
FIG. 5 is a diagram for explaining a functional pixel pattern (rectangular arrangement) and a scanning method of multipoint scanning in the confocal microscope according to the embodiment of the present invention.
FIG. 6 is a diagram illustrating a functional pixel pattern (alternate arrangement) and a scanning method of multipoint scanning in the confocal microscope according to the embodiment of the present invention.
FIG. 7 is a view for explaining an alternate arrangement of all effective screens in the vertical and horizontal proportions in the confocal microscope according to the embodiment of the present invention.
FIG. 8 is a diagram for explaining a rectangular arrangement of functional pixels of a 2 × 2 pixel configuration in the confocal microscope according to the embodiment of the present invention, which is proportional to the vertical and horizontal directions of the entire effective screen.
FIG. 9 is a diagram for explaining a square arrangement with an interval of 5 in functional pixels having a 3 × 3 pixel configuration in the confocal microscope according to the embodiment of the present invention.
FIG. 10 is a diagram for explaining a hexagonal arrangement of a functional pixel having a 5 × 5 pixel configuration and an interval of 6 in the confocal microscope according to the embodiment of the present invention.
FIG. 11 is a diagram showing another example of the arrangement of functional pixels in the confocal microscope according to the embodiment of the present invention.
FIG. 12 is a diagram for explaining a functional pixel interval and crosstalk due to an out-of-focus image in the confocal microscope according to the embodiment of the present invention.
FIG. 13 is a diagram illustrating the relationship between the telecentricity of the confocal optical system and the shift of the image position in the confocal microscope according to the embodiment of the present invention.
FIG. 14 is a diagram for explaining an image shift between the two-dimensional shutter array and the two-dimensional detector in the confocal microscope according to the embodiment of the present invention.
FIG. 15 is a configuration diagram of a confocal microscope according to another embodiment of the present invention.
FIG. 16 is a configuration diagram of a confocal microscope according to another embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Illumination system, 2 ... Light transmission system, 3 ... Light reception system, 4 ... Control system, 11 ... Light source, 12 ... Light color filter, 13 ... Condensing lens, 21 ... Two-dimensional shutter array, 22 ... Collimator lens, 23 ... Beam splitter, 24 objective lens, 25 sample, 26 excitation filter, 27 dichroic mirror, 28 polarizer, 29 Nomarski prism, 31 second objective lens, 32 two-dimensional detector, 33 absorption filter , 34: Wave plate, 35: Analyzer, 40: Control device, 211: Functional emission pixel (shutter), 212: Idle emission pixel (shutter), 321: Functional light receiving pixel (sensor), 322: Idle light receiving pixel (sensor) ).

Claims

A plurality of emission pixels capable of emitting light, emission means arranged two-dimensionally,
Light receiving means having a plurality of light receiving pixels corresponding to the plurality of emission pixels of the emission means,
A confocal optical system that irradiates a sample with light emitted from the emission unit and condenses return light from the sample surface on the light receiving unit,
One of the plurality of emission pixels or a plurality of adjacent pixels is set as a function emission pixel, and at least two of the plurality of function emission pixels are set at non-adjacent positions. A light receiving pixel corresponding to the function emitting pixel is set as a function light receiving pixel, and a process of emitting light from the function emitting pixel and receiving the return light at the function light receiving pixel is performed by a different one of the plurality of emission pixels. A confocal microscope comprising: a control unit that obtains an image of the sample by performing a plurality of times on emission pixels or functional emission pixels set in different combinations of emission pixels.

The confocal microscope according to claim 1, wherein
An interval between the at least two functional emission pixels and an interval between at least two functional light receiving pixels corresponding to the at least two functional emission pixels are more than 10 times and less than 100 times the diameter of each Airy disk. Confocal microscope.

The confocal microscope according to claim 1, wherein
The control means sets the interval between the at least two functional emission pixels and the interval between at least two functional light receiving pixels corresponding to the at least two functional emission pixels to be larger as the step of the unevenness on the sample surface is larger. Features confocal microscope.

The confocal microscope according to claim 1, wherein
The control means sets an interval between the at least two functional emission pixels and an interval in pixel units of at least two functional light receiving pixels corresponding to the at least two functional emission pixels,
(Step of unevenness of sample) / (2.44 × depth of focus)
A confocal microscope characterized by being set larger.

The confocal microscope according to claim 1, wherein
The confocal microscope, wherein the control means changes the number of pixels constituting the functional emission pixels and the functional light receiving pixels according to an Airy disk diameter depending on a pupil diameter of an objective lens used.

The confocal microscope according to claim 1, wherein
The control unit may perform a process of emitting light from the functional emission pixel and causing the functional light receiving pixel to receive the return light for a plurality of functional emission pixels set by different combinations of emission pixels among the plurality of emission pixels. When performing the number of times, the distance between the centers of the functional emission pixels set a plurality of times is set to one emission pixel or less than the number of vertical or horizontal emission pixels constituting the functional emission pixel. A confocal microscope characterized in that:

The confocal microscope according to claim 1, wherein
The control means may arrange the plurality of functional emission pixels and the light receiving function pixels in a square arrangement, a hexagonal arrangement, a rectangular arrangement having a ratio equal to the aspect ratio of the entire effective screen, or an alternating arrangement having a ratio equal to the aspect ratio of the entire effective screen. And a confocal microscope.

The confocal microscope according to claim 1, wherein
The control unit performs a process of emitting light from the functional emission pixel and causing the functional light receiving pixel to receive the return light a plurality of times for functional emission pixels set by different emission pixels among the plurality of emission pixels. A confocal microscope characterized in that when obtaining an image of the sample, an image of the sample is formed by thinning scanning and interpolation.

The confocal microscope according to claim 1, wherein
The confocal microscope, wherein the confocal optical system has a collimator lens, a first objective lens, and a second objective lens in which a maximum distortion and a maximum chromatic aberration of magnification are corrected to 1/5 or less of the light receiving pixel size. .

The confocal microscope according to claim 9, wherein
The first objective lens and the second objective lens are characterized in that the maximum chief ray inclination angle is corrected to an angle such that the displacement amount within the depth of focus is within 1/5 of the size of the light receiving pixel. And confocal microscope.