JP4311952B2 - 3D coordinate measurement method - Google Patents

Description

ここで、（ａ）式の第１項は球面項、第２項は自由曲面項である。
【００３１】
球面項中、
ｃ：頂点の曲率
ｋ：コーニック定数（円錐定数）
ｒ＝√（Ｘ² ＋Ｙ² ）
である。
【００３２】
自由曲面項は、

ただし、Ｃ_j （ｊは２以上の整数）は係数である。
【００３３】
上記自由曲面は、一般的には、Ｘ−Ｚ面、Ｙ−Ｚ面共に対称面を持つことはないが、Ｘの奇数次項を全て０にすることによって、Ｙ−Ｚ面と平行な対称面が１つだけ存在する自由曲面となる。また、Ｙの奇数次項を全て０にすることによって、Ｘ−Ｚ面と平行な対称面が１つだけ存在する自由曲面となる。
【００３４】
次に、複数の面の境界からなるエッジを、そのエッジを構成する面毎に測定し、それらを組み合わせることで、エッジを構成する複数の面の相対的な偏心を算出する方法について説明する。ここでは、例えば、自由曲面や、シリンドリカル面、アナモルフィック面、非球面、球面が複数面相互に偏心して配置されることで構成されるプリズム等の光学素子の各面の設計座標系（ローカル座標系）の相対的な位置関係を求める場合を示す。
【００３５】
図７を参照にして、被検物（プリズム）Ｏの１つのエッジＥを共有する被検面Ｓ１、Ｓ２の偏心量を算出する方法について説明する。図７（ａ）は被検物Ｏの被検面Ｓ１に対して光プローブ２を走査することにより、その面の形状、エッジの形状を測定する様子を示す斜視図、図７（ｂ）は被検面Ｓ１の上面図、図７（ｃ）は被検面Ｓ２に対して光プローブ２を走査することにより、その面の形状、エッジの形状を測定する様子を示す斜視図、図７（ｄ）は被検面Ｓ２の上面図である。
【００３６】
まず、図６で示したように、各被検面Ｓ１、Ｓ２毎にエッジＥの形状測定値と有効エリアＢ１、Ｂ２内の面形状測定値とを同じ非接触３次元測定機を用いて同じ測定座標系で測定し、その面形状データを各被検面Ｓ１、Ｓ２の設計値に当てはめるフィッティングを行い、その座標変換をエッジＥの測定データにも行うことで、被検面Ｓ１、Ｓ２を表すローカル座標系Ａ１、Ａ２に対するエッジＥの位置を求める。
【００３７】
さらに、各面毎に測定したエッジＥの各データが一致するように座標変換を行う。このとき、一方の面のエッジ測定値を固定とし、他方のエッジ測定値を偏心、シフトさせることで、面エッジが重なるようにしてもよい。また、解析には両エッジの測定値が略同一曲線上に乗るように最小２乗法を利用してもよい。
【００３８】
この解析より得られた座標変換を、被検面Ｓ１、Ｓ２を表すローカル座標系Ａ１、Ａ２に対しても実行することで、図８に示すような、複数の面Ｓ１、Ｓ２におけるローカル座標系Ａ１、Ａ２の相対的な位置関係を明確にすることが可能となる。なお、図８では、エッジＥを構成する被検面が２面のみの例を示したが、これに限定するものではなく、被検面が３面以上のローカル座標系の相対位置も同様の手法により求めることが可能となる。これらの手法により、複数の面における相対的な偏心量が分かるため、光学性能劣化の要因解明、修正等に反映させることが可能となる。
【００３９】
次に、図９に示すように、共通のエッジＥを共有する２つの被検面Ｓ１、Ｓ２の偏心量を、被検物Ｏの素子に固定した基準となる素子（基準ブロック）Ｄを用いて、算出する方法について説明する。図９（ａ）は被検物Ｏの被検面Ｓ１と基準ブロックＤの１つの基準面Ｈに対して光プローブ２を走査することにより、被検面Ｓ１の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図９（ｂ）は被検面Ｓ１の上面図、図９（ｃ）は被検面Ｓ２と基準ブロックＤの他の基準面Ｈに対して光プローブ２を走査することにより、被検面Ｓ２の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図９（ｄ）は被検面Ｓ２の上面図である。
【００４０】
被検物Ｏは基準ブロックＤと保持具Ｃで一体に固定して保持されており、基準ブロックＤの各基準面Ｈとそれに対応した各被検面Ｓ１、Ｓ２の形状をそれぞれ同じ非接触３次元測定機を用いて同じ測定座標系で測定できるようにする。
【００４１】
なお、各被検面Ｓ１、Ｓ２の面形状データは、それぞれ各被検面Ｓ１、Ｓ２の設計値に当てはめるフィッティングを行うことで、被検面Ｓ１、Ｓ２を表すローカル座標系ローカル座標系Ａ１、Ａ２の位置を求めておく。
【００４２】
また、ローカル座標系算出の際には、各被検面の設計形状を座標変換して、面形状データと座標変換した各設計形状を重ね合わせた際に、残差が最小となるような座標変換パラメータを求める最適化の解析を行ってもよい。
【００４３】
基準ブロックＤの形状は、被検物Ｏの各被検面Ｓ１、Ｓ２の測定と同時に少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点の位置を測定できるように構成することが望ましい。このとき、測定に用いた少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点は、それらの測定結果より基準ブロックＤの各基準面Ｈの位置を定義できるようなものとする必要がある。さらに、少なくとも被検物Ｏの各測定面と同時に測定するエッジｅの角度や長さは、予め高精度に測定しておき、既知のものとしておき、これらの既知のデータから測定値で定義される各基準面Ｈの相対的な位置が分かるようにしておく。
【００４４】
なお、基準ブロックＤの形状は、例えば、被検物Ｏの形状に合わせて被検面Ｓ１、Ｓ２を平面で近似した構成の多面体の形状であってもよい。このとき、各被検面Ｓ１、Ｓ２の設計値のローカル座標系Ａ１、Ａ２で定義される平面で構成した多面体としてもよい。基準ブロックＤを平面で構成する場合、測定が高精度かつ簡略化でき、解析も他の曲面に対して容易となる。また、このような形状にすることにより、基準ブロックＤの各基準面Ｈとそれに対応した各被検面Ｓ１、Ｓ２を略水平に保持することが可能となり、同じ測定座標系を用いて、それぞれ容易に測定することが可能となる。
【００４５】
次に、測定方法について説明する。まず、図９で示すように、各被検面Ｓ１、Ｓ２毎に基準ブロックＤのエッジｅの形状測定値と、被検物Ｏの有効エリアＢ１、Ｂ２内の面形状測定値とを、同じ非接触３次元測定機を用いて同じ測定座標系で測定する。また、後述するように、有効エリア以外の領域に、被検面上のローカル座標系で示した場合の座標位置が既知であるマークｖが加工されている場合には、同様に、同じ測定座標系で測定する。
【００４６】
なお、例えば測定機の関係で、被検面Ｓ１、Ｓ２を上に向ける必要がある場合、図９（ａ）〜（ｄ）に示すように、各面の測定毎に被検物Ｏを回転させて保持する必要がある。そのため、図示しない回転機構を用いて、被検物Ｏと基準ブロックＤを所望の方向に回転させて固定することになる。このとき、被検物Ｏと基準ブロックＤは、回転によっても相対的な位置関係が変化しないようにする。
【００４７】
基準ブロックＤは、既知の測定値に対応させて、基準ブロックＤの測定値より基準面位置を求めることができるような項目の測定を行う。図９では、各基準面Ｈ毎に２つのエッジｅを測定した例を示している。これらのデータから、基準ブロックＤの基準面Ｈの相対位置を求めることが可能となる。
【００４８】
例えば、被検面Ｓ１とＳ２の各基準面Ｈのエッジの測定で、測定するエッジは、基準ブロックＤを構成する基準面の境界となる稜線を含むものとし、さらに異なるエッジ部を測定することで、各基準面毎に稜線の位置と基準面を表す平面の位置を求めるようにしてもよい。ここで、稜線の位置とは、稜線が示す基準面の辺の少なくとも１つの頂点の位置と、稜線を表す直線式を導出した結果を意味する。頂点を算出するためには、基準面内で頂点をなす２つのエッジの形状の測定結果から交点を求めてもよい。また、例えば、基準面の頂点との位置関係が既知である点を測定できるようにしておき、その点の測定結果と稜線となるエッジの測定結果から、同様に稜線の位置と基準面を表す平面の位置を求めるようにしてもよい。このとき、解析にはエッジの測定値が略直線上になるように、最小２乗法等を利用してもよい。さらに、導出した平面と基準ブロックＤの既知の測定値（辺の長さや辺と辺の角度）を対応させて、各基準面の位置を求めることが可能となる。ただし、各基準面位置として、平面の位置と稜線の位置が分かれば、後述する各被検面Ｓ１、Ｓ２の相対値の導出は可能となる。
【００４９】
さらに、同じ測定座標系で被検面Ｓ１、Ｓ２の形状を測定することにより、基準面Ｈと被検面Ｓ１、Ｓ２の相対的な位置を各被検面Ｓ１、Ｓ２毎に把握することが可能となる。
【００５０】
なお、後述するように、有効エリア以外の領域に、被検面上のローカル座標系で示した場合の座標位置が既知であるマークｖが加工されている場合には、面形状と同様に、マーク位置を測定することで、基準面Ｈと被検面Ｓ１、Ｓ２上にあるマークｖの相対的な位置を、各被検面Ｓ１、Ｓ２毎に把握することが可能となる。
【００５１】
さらに、基準ブロックＤを構成する基準面Ｈの既知のデータに対応させて、各基準面Ｈの測定値により求められた各面の位置を変換し、各面の相対的な位置関係を合わせる。このとき、シフトや回転といった座標変換を行う必要がある。
【００５２】
例えば、同一の稜線を含む各基準面Ｈの測定結果からそれぞれ導出した稜線の位置が空間的に一致するように座標変換を行い、これにより空間的に一致した稜線を回転軸として、稜線を含む各平面を、基準素子の既知の値（基準素子で同じ稜線を構成する各基準面のなす角度) と対応するように、回転の座標変換を行うパラメータを算出してもよい。これにより、基準ブロックＤの既知の角度を用いて、各基準面Ｈの相対位置を求めることが可能となる。
【００５３】
これらの変換を各基準面と同じ測定座標系で測定した各被検面Ｓ１、Ｓ２の形状測定データに対しても行うことで、各被検面Ｓ１、Ｓ２の相対的な位置を算出することが可能となる。
【００５４】
また、後述するように、有効エリア以外の領域に、被検面上のローカル座標系で示した場合に座標位置が既知であるマークｖが加工されている場合には、面形状と同様に、各被検面と同じ測定座標系で各マーク位置を測定し、それらのマーク位置に対しても同じ変換を行うことで、被検面Ｓ１、Ｓ２上にある各マークｖの相対的な位置を算出することが可能となる。
【００５５】
この解析により得られた座標変換を、各被検面Ｓ１、Ｓ２のローカル座標系Ａ１、Ａ２にも実行することで、図１０に示すような、複数の面Ｓ１、Ｓ２におけるローカル座標系Ａ１、Ａ２の相対的な位置関係を明確にすることが可能となる。なお、図１０では、エッジＥを構成する２面のみの例を示したが、これに限定するものではなく、３面以上のローカル座標系の相対位置も同様の手法により求めることが可能となる。また、本実施例では、１つの被検面に対して基準ブロックＤの１つの基準面を同時に測定したが、被検物の形状によっては２つ以上の被検物に対して１つの基準面を同時に測定するようにしてもよい。これらの手法により、複数の面における相対的な偏心量が分かるため、光学性能劣化の要因解明、修正等に反映させることが可能となる。
【００５６】
また、本手法の結果を基に製造を実施することで、精度の高い光学素子を作製することが可能となる。
【００５７】
図１１〜図１４を参照にして、枠Ｗに光学素子（例えばプリズム）が複数配置された被検物Ｏの、各光学素子の偏心量を、被検物Ｏを保持する枠Ｗに固定した基準となる素子（基準ブロック）Ｄを用いて算出する方法について説明する。このとき、被検物Ｏは、ある方向において、各光学素子を構成する少なくとも１面の面形状が測定できるようにする。図１１〜図１４では、プリズムＰ１とプリズムＰ２が枠Ｗ内に配置され、プリズムＰ１を構成する面Ｓ１−１とプリズムＰ２を構成するＳ２−２が外側に出ており、光プローブ２を照射することで各面の測定が可能となっている。
【００５８】
図１１は、プリズムＰ１の被検面Ｓ１−１と基準ブロックＤの１つの基準面Ｈに対して光プローブ２を走査することにより、被検面Ｓ１−１の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図１２は、被検面Ｓ１−１の上面図、図１３は、プリズムＰ２の被検面Ｓ２−１と基準ブロックＤの他の基準面Ｈに対して光プローブ２を走査することにより、被検面Ｓ２−１の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図１４は、被検面Ｓ２−１の上面図である。
【００５９】
被検物Ｏは基準ブロックＤと保持具Ｃで一体に固定して保持されており、基準ブロックＤの各基準面Ｈとそれに対応した各被検面Ｓ１−１、Ｓ２−１の形状をそれぞれ非接触３次元測定機を用いて同じ測定座標系で測定できるようにする。なお、各被検面Ｓ１−１、Ｓ２−１の面形状データは、それぞれ各被検面Ｓ１−１、Ｓ２−１の設計値に当てはめるフィッティングを行うことで、被検面Ｓ１−１、Ｓ２−１を表すローカル座標系Ａ１−１、Ａ２−１の位置を求めておく。
【００６０】
また、ローカル座標系算出の際には、各被検面の設計形状を座標変換して、面形状データと座標変換した各設計形状を重ね合わせた際に、残差が最小となるような座標変換パラメータを求める最適化の解析行ってもよい。
【００６１】
基準面Ｈの形状は、被検物ＯのプリズムＰ１、Ｐ２の各被検面Ｓ１−１、Ｓ２−１の測定と同時に、少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点の位置を測定できるように構成することが望ましい。このとき、測定に用いた少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点は、それらの測定結果より基準ブロックＤの各基準面Ｈの位置を定義できるようなものとする必要がある。さらに、少なくとも各測定面Ｈと同時に測定するエッジｅの角度や長さは、予め高精度に測定しておき、これらの既知のデータから測定値で定義される各基準面Ｈの相対的な位置が分かるようにしておく。
【００６２】
なお、基準ブロックＤの形状は、例えば、被検物のプリズムＰ１、Ｐ２の形状に合わせて各被検面Ｓ１−１、Ｓ２−１を平面で近似した面を有する多面体の形状であってもよい。このとき、各被検面Ｓ１−１、Ｓ２−１の設計値のローカル座標系Ａ１−１、Ａ２−１で定義される平面を有する多面体の形状であってもよい。基準ブロックＤを平面で構成する場合、測定が高精度かつ簡略化でき、解析も他の曲面に対して容易となる。また、材質としては、ガラスや金属等であってもよい。これらの材質で製作すると、高い面精度での加工が可能となるため、基準面の角度を高精度に測定することができる。また、鋭利なエッジ形状が形成できるため、基準面の高精度な測定が可能となる。
【００６３】
次に、測定方法について説明する。まず、図１１〜図１４で示すように、各被検面Ｓ１−１、Ｓ２−１毎に基準ブロックＤのエッジｅの形状と、被検物ＯであるプリズムＰ１、Ｐ２の各被検面Ｓ１−１とＳ２−１の有効エリアＢ１−１とＢ２−１内の面形状を、同じ非接触３次元測定機を用いて同じ測定座標系で測定する。このとき、例えば、測定機の関係で、光プローブ２の入射する方向に対して被検面Ｓ１−１、Ｓ２−１が略垂直になるように、被検面Ｓ１−１、Ｓ２−１を上に向ける必要がある場合、図１１〜図１４に示すように、各面の測定毎に被検物Ｏを回転して保持する必要がある。そのため、図示しない回転機構を用いて、被検物Ｏと基準ブロックＤを所望の方向に回転させて固定することになる。このとき、被検物Ｏと基準ブロックＤは、回転によっても相対的な位置関係が変化しないようにする。
【００６４】
このとき、基準ブロックＤを、各被検面Ｓ１−１、Ｓ２−１の設計値のローカル座標系Ａ１−１、Ａ２−１で定義される平面を有する多面体の形状にしておくと、被検物Ｏの各被検面Ｓ１−１、Ｓ２−１と基準ブロックＤの基準面の測定の際に、被検物Ｏの各被検面Ｓ１−１、Ｓ２−１と基準ブロックＤの基準面が、同時に光プローブ２の入射する方向に対して略垂直になるように配置することが可能となるため、測定を容易に実行することができる。
【００６５】
基準ブロックＤは、例えば各基準面の辺の長さといった既知の測定値に対応させることで、基準ブロックＤの測定値より基準面位置を求めることができるような項目の測定を行う。図１１〜図１４では、各基準面Ｈ毎に２つのエッジを測定した例を示している。これらの測定データから、基準ブロックＤの基準面Ｈの位置を求めることが可能となる。
【００６６】
なお、基準面Ｈの位置の導出には、図９で示した例と同様の手法を用いることができる。これにより、基準ブロックＤの各基準面Ｈの位置を求めることが可能となる。さらに、同じ測定座標系で被検面Ｓ１−１、Ｓ２−１の形状を測定することにより、基準面Ｈと被検面Ｓ１−１、Ｓ２−１の相対的な位置を各被検面Ｓ１−１、Ｓ２−１毎に把握することが可能となる。
【００６７】
さらに、基準ブロックＤを構成する基準面Ｈの既知のデータに対応させて、各基準面Ｈの測定値により求められた各面の位置を変換し、各面の相対的な位置関係を合わせる。このとき、シフトや回転といった座標変換を行う必要がある。
【００６８】
なお、各面の相対的な位置を求めるのには、図９で示した例と同様の手法を用いることができる。これらの変換を各基準面と同じ測定座標系で測定した各被検面Ｓ１−１、Ｓ２−１に対しても行うことで、被検面Ｓ１−１、Ｓ２−１の相対的な位置を把握することが可能となる。
【００６９】
この解析により得られた座標変換を、各被検面Ｓ１−１、Ｓ２−１の各ローカル座標系Ａ１−１、Ａ２−１にも実施することで、図１５に示すような、異なる光学素子（プリズムＰ１、Ｐ２）を構成する複数の面Ｓ１−１、Ｓ２−１におけるローカル座標系Ａ１−１、Ａ２−１の相対的な位置関係を求めることが可能となる。
【００７０】
なお、図１１〜図１４では、２個のプリズムＰ１、Ｐ２から構成される例を示したが、これに限定されるものではなく、３個以上の光学素子の相対位置も同様の手法により求めることが可能となる。また、光学素子（プリズムＰ１、Ｐ２）について、被検面を１面とした例を示したが、これに限定されるものではなく、各素子について、複数の面を測定してもよく、これによりより多くの偏心及び寸法の情報を得ることが可能となる。この際には、基準ブロックの基準面の数を必要に応じて増やしてもよい。また、被検物として異なるプリズムの相対位置を求める例を示したが、被検物はこれに限定されるものではなく、レンズ、光学素子の金型、絞り、枠Ｗといったものでもよい。例えば、枠の場合、被検面の面形状を測定してもよいが、測定値から枠の位置が決定できるものを測定すればよく、例えば、凹凸、マークといったものと基準ブロックＤの基準面Ｈを同じ測定座標系で測定してもよい。
【００７１】
また、枠Ｗに組み込む前に、各プリズムの各面のローカル座標の相対位置を、例えば図９で示した手法により求めておけば、各プリズム毎にプリズムを構成する各面のローカル座標系の相対位置が既知であるため、それらの結果を図１５で示した各プリズムＰ１、Ｐ２の被検面Ｓ１−１、Ｓ２−１をローカル座標系Ａ１−１、Ａ２−１の位置に対応させることで、図１６に示すように、プリズムＰ１を構成するＳ１−１以外の面（Ｓ１−２、Ｓ１−３）及びプリズムＰ２を構成するＳ２−１以外の面（Ｓ２−２、Ｓ２−３）の各面のローカル座標系（Ａ１−２、Ａ１−３、Ａ２−２、Ａ２−３）の相対位置を求めることができる。これらの手法により、複数の素子における相対的な偏心量が分かるため、光学性能劣化の要因解明、修正等に反映させることが可能となる。また、本手法の結果を基に製造を実施することで、精度の高い光学系を作製することが可能となる。
【００７２】
次に、図１７を参照にして、例えば非球面レンズのような被検物Ｏに対して、表裏に配置された被検面のＳ３−１とＳ３−２の偏心量を、被検物Ｏに固定した基準となる素子（基準ブロック）Ｄを用いて算出する方法について説明する。図１７（ａ）は非球面レンズＯの被検面Ｓ３−１と基準ブロックＤの１つの基準面Ｈに対して光プローブ２を走査することにより、被検面のＳ３−１の形状、基準面Ｈのエッジｅの形状を測定する様子を示す横から見た図、図１７（ｂ）は被検面Ｓ３−１の斜視図、図１７（ｃ）は被検面Ｓ３−２と基準ブロックＤの他の基準面Ｈに対して光プローブ２を走査することにより、被検面のＳ３−２の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図である。
【００７３】
被検物Ｏは基準ブロックＨと保持具Ｃで一体に固定して保持されており、基準ブロックＤの各基準面Ｈとそれに対応した各被検面Ｓ３−１、Ｓ３−２の形状をそれぞれ非接触３次元測定機を用いて同じ測定座標系で測定できるようにする。なお、各被検面Ｓ３−１、Ｓ３−２の面形状データは、それぞれ各被検面Ｓ３−１、Ｓ３−２の設計値に当てはめるフィッティングを行うことで、被検面Ｓ３−１、Ｓ３−２を表すローカル座標系Ａ１−１、Ａ２−１の位置を求めておく。
【００７４】
基準ブロックＨの形状は、被検物Ｏの非球面レンズの各被検面Ｓ３−１、Ｓ３−２の測定と同時に少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点の位置を測定できるように構成することが望ましい。このとき、測定に用いた少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点は、それらの測定結果より基準ブロックＤの各基準面にＨの位置を定義できるようなものとする必要がある。さらに、少なくとも各測定面Ｈと同時に測定するエッジｅの角度や長さは、予め高精度に測定しておき、これらの既知のデータから測定値で定義される各基準面Ｈの相対的な位置が分かるようにしておく。
【００７５】
なお、基準ブロックＤの形状は、例えば、被検物Ｏの各被検面Ｓ３−１、Ｓ３−２の厚さ程度の直方体であってもよい。このとき、対向する平面を基準面として、各基準面で２つのエッジを測定してもよい。このとき、基準ブロックＤの各辺の長さ、各基準面間の平行度や基準面に隣合う面との角度（直交度）、基準面間の高さ等を高精度に測定しておくことが望ましい。
【００７６】
次に、測定方法について説明する。まず、図１７に示すように、各被検面Ｓ３−１、Ｓ３−２毎に基準ブロックＤのエッジｅの形状と、被検物Ｏの各被検面Ｓ３−１、Ｓ３−２の有効エリアの面形状を、非接触３次元測定機を用いて同じ測定座標系で測定する。このとき、例えば測定機の関係で、光プローブ２の入射する方向に対して被検面が略垂直になるように、被検面Ｓ３−１、Ｓ３−２を上に向ける必要がある場合、図１７（ａ）〜（ｃ）に示すように、各面の測定毎に被検物Ｏを略１８０°回転して保持する必要がある。このとき、被検物Ｏと基準ブロックＤは、回転によっても相対的な位置関係が変化しないようにする。
【００７７】
基準ブロックＤは、例えば各基準面の辺の長さといった既知の測定値に対応させることで、基準ブロックＤの測定値より基準面位置を求めることができるような項目の測定を行う。図１７では、各基準面Ｈ毎に２つのエッジを測定した例を示している。これらの測定データから、基準ブロックＤの基準面Ｈの位置を求めることが可能となる。なお、基準面Ｈの位置の導出には、図９で示した例と同様の手法を用いることができる。これにより、基準ブロックＤの各基準面Ｈの位置を求めることが可能となる。さらに、同じ測定座標系で被検面Ｓ３−１、Ｓ３−２の形状を測定することにより、基準面Ｈと被検面Ｓ３−１、Ｓ３−２の相対的な位置を各被検面Ｓ３−１、Ｓ３−２毎に把握することが可能となる。
【００７８】
さらに、基準ブロックＤを構成する基準面Ｈの既知のデータに対応させて、各基準面Ｈの測定値により求められた各面の位置を変換し、各面の相対的な位置関係を合わせる。基準ブロックＤが略直方体の場合、例えば、測定により算出した各基準面の頂点の座標と各基準面の傾きを、既知のデータの頂点の相対位置と平行度のデータと対応させてもよい。なお、図１７では、基準面の境界となる稜線のエッジの形状は測定していないが、各基準面のエッジの相対的な位置関係が既知であれば、解析は可能となる。これらの変換を各基準面と同じ測定座標系で測定した各被検面Ｓ３−１、Ｓ３−２に対しても行うことで、被検面Ｓ３−１、Ｓ３−２の相対的な位置を把握することが可能となる。また、対向する基準面に加えて、両基準面とそれぞれ接する基準面の相対位置を測定し、それらの結果を、既知である基準ブロックの各面の角度に対応させて、対向する基準面の相対位置を導出する解析を行ってもよい。
【００７９】
この解析により得られた座標変換を、各被検面Ｓ３−１、Ｓ３−２の各ローカル座標系Ａ３−１、Ａ３−２にも実施することで、図１７（ｄ）に示すように、非球面レンズを構成する面Ｓ３−１、Ｓ３−２におけるローカル座標系Ａ３−１、Ａ３−２の相対的な位置関係を明確にすることが可能となる。また、解析により各被検面の相対位置が分かるので、偏心の他に、レンズの肉厚等の情報も得ることが可能となる。さらに、これらの情報を基に製造を実施することで精度の高い光学素子を作製することが可能となる。
【００８０】
被検面の形状が非球面の場合、被検面の面形状データから設計値に当てはめるフィッティングを行うことでローカル座標の位置を求めることができるが、被検面の形状が球面の場合は、同手法を用いてもローカル座標系の位置を決定することができない。そのような場合は、被検面上に、被検面の形状の測定時に同時に測定することで、ローカル座標系の位置を求めることが可能となるようなマークＭを付けておいてもよい。この場合、ローカル座標系に対するマークＭの相対位置は既知としておくことで、複数のマークの測定結果から、ローカル座標系の位置の導出が可能となる。なお、マークの形状は、上記解析が可能であれば形状を限定するものではなく、例えば、図１８（ａ）に示すような十字や、図１８（ｂ）に示す円柱であってもよい。また、外形から上記解析が可能であれば、マーク等を加工する必要はない。例えば、図１８（ｃ）に示すような、小判型の外形をしている場合、外形の直線部の形状や、円弧部との頂点の位置等からローカル座標系の相対位置が導出できれば、それらの測定結果から被検面のローカル座標系を導出する解析を行ってもよい。
【００８１】
また、被検面の形状が球面の場合、例えば、面形状データを球でフィッティングすることで球心位置を算出することは可能なので、各面の相対的な球心位置を求めることも可能である。その際に、球面の被検面の測定において、例えば図１８に示すようなマークや外形といったものを同じ座標系で測定しておくことが望ましい。
【００８２】
なお、外形の測定では、図１８（ｃ）に示すような小判型に限定するものではなく、通常の円形の外形であってもよい。その場合は、外形の測定結果を円でフィッティングすることにより、外形中心の位置を求めることが可能となる。それによって、例えば、マークの位置や外形中心といった基準位置に対する各面の球心の相対位置を求めることが可能となる。また、同様に非球面を含んだ場合にも、例えば、非球面の面形状データを球面でフィッティングすることで、ローカル座標系の他に、近似球心位置を求めることも可能である。このように本手法を用いることで、回転対称な光学面を有する被検物に対して、偏心の測定を行うことが可能となる。
【００８３】
次に、図２０を参照にして、図１９に示すような枠Ｗに光学素子（例えば、レンズ、プリズム）が複数配置された被検物Ｏの各光学素子の偏心量を、被検物Ｏの枠Ｗに固定した基準となる素子（基準ブロック）Ｄを用いて算出する方法について説明する。このとき、被検物Ｏは、プリズムによって光軸が折り曲げられた光学系となっている。なお、被検物Ｏは、ある方向において複数の面の面形状が測定できるようにする。そのため、枠Ｗは、被検面測定の際に、枠Ｗと測定機を構成する部材、例えば図示しない光プローブ２を照射するレンズが衝突しないような構成とすることが望ましい。
【００８４】
図１９では、第１のレンズ群Ｇ１とプリズムＰ３と第２のレンズ群Ｇ２が枠Ｗ内に配置され、第１のレンズ群Ｇ１のレンズＬ１−１を形成する面ＳＬ１−１とプリズムＰ３を形成する面ＳＰ３−２と、第２のレンズ群Ｇ２のレンズＬ２−５を形成する面ＳＬ２−５が外側に出ており、光プローブ２を照射することで各面の測定が可能となっている。
【００８５】
図２０（ａ）は、被検面ＳＬ１−１と基準ブロックＤの１つの基準面Ｈに対して光プローブ２を走査することにより、被検面ＳＬ１−１の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図２０（ｂ）は、被検面ＳＰ３−２と基準ブロックＤの他の基準面Ｈに対して光プローブ２を走査することにより、被検面のＳＰ３−２の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図２０（ｃ）は、被検面ＳＬ２−５と基準ブロックＤの他の基準面Ｈに対して光プローブ２を走査することにより、被検面ＳＬ２−５の形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図である。
【００８６】
被検物Ｏは基準ブロックＨと保持具Ｃで一体に固定して保持されており、基準ブロックＤの各基準面Ｈとそれに対応した各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の形状をそれぞれ非接触３次元測定機を用いて同じ測定座標系で測定できるようにする。なお、各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の面形状データは、それぞれ各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の設計値に当てはめるフィッティングを行うことで、被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５を表すローカル座標系ＡＬ１−１−１、ＡＰ３−２、ＡＬ２−５−２の位置を求めておく。
【００８７】
また、ローカル座標系算出の際には、各被検面の設計形状を座標変換して、面形状データと座標変換した各設計形状を重ね合わせた際に、残差が最小となるような座標変換パラメータを求める最適化の解析を行ってもよい。
【００８８】
基準ブロックＤの形状は、被検物ＯのレンズＬ１−１、Ｌ２−５、プリズムＰ３の各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の測定と同時に少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点の位置を測定できるように構成することが望ましい。このとき、測定に用いた少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点は、それらの測定結果より基準ブロックＤの各基準面にＨの位置を定義できるようなものとする必要がある。さらに、少なくとも各測定面Ｈと同時に測定するエッジｅの角度や長さは、予め高精度に測定しておき、これらの既知のデータから測定値で定義される各基準面Ｈの相対的な位置が分かるようにしておく。
【００８９】
なお、基準ブロックＤの形状は、例えば、被検物Ｏの形状に合わせて各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５を平面で近似した面を有する多面体の形状であってもよい。このとき、各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の設計値のローカル座標系ＡＬ１−１−１、ＡＰ３−２、ＡＬ２−５−２で定義される平面を有する多面体の形状であってもよい。
【００９０】
次に、測定方法について説明する。まず、図２０で示すように、各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５毎に基準ブロックＤのエッジｅの形状と、被検物Ｏの各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の有効エリア内の面形状とを、非接触３次元測定機を用いて同じ測定座標系で測定する。このとき、例えば測定機の関係で、光プローブ２の入射する方向に対して被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５が略垂直になるように、被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５を上に向ける必要がある場合、図２０（ａ）〜（ｃ）に示すように、各面の測定毎に被検物Ｏを回転して保持する必要がある。そのため、図示しない回転機構を用いて、被検物Ｏと基準ブロックＤを所望の方向に回転させて固定することになる。このとき、被検物Ｏと基準ブロックＤは、回転によっても相対的な位置関係が変化しないようにする。
【００９１】
基準ブロックＤは、例えば各基準面の辺の長さといった既知の測定値に対応させることで基準ブロックＤの測定値より基準面位置を求めることができるような項目の測定を行う。図２０では、各基準面Ｈ毎に２つのエッジを測定した例を示している。これらの測定データから、基準ブロックＤの基準面Ｈの位置を求めることが可能となる。なお、基準面Ｈの位置の導出には、図９で示した例と同様の手法を用いることができる。これにより、基準ブロックＤの各基準面Ｈの位置を求めることが可能となる。さらに、同じ測定座標系で被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の形状を測定することにより、基準面Ｈと被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の相対的な位置を各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５毎に把握することが可能となる。
【００９２】
さらに、基準ブロックＤを構成する基準面Ｈの既知のデータに対応させて、各基準面Ｈの測定値により求められた各面の位置を変換し、各面の相対的な位置関係を合わせる。このとき、シフトや回転といった座標変換を行う必要がある。なお、各面の相対的な位置を求めるのには、図９で示した例と同様の手法を用いることができる。これらの変換を各基準面と同じ測定座標系で測定した各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５に対しても行うことで、被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の相対的な位置を把握することが可能となる。
【００９３】
この解析により得られた座標変換を、各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５の各ローカル座標系ＡＬ１−１−１、ＡＰ３−２、ＡＬ２−５−２にも実施することで、図２２に示すような、複数の面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５におけるローカル座標系ＡＬ１−１−１、ＡＰ３−２、ＡＬ２−５−２の相対的な位置関係を求めることが可能となる。
【００９４】
また、枠Ｗに組み込む前に、被検物内に配置された各光学素子の各面のローカル座標の相対位置を、例えば各群やプリズム毎に求めておくことが望ましい。図２１（ａ）〜（ｃ）は、被検物を構成するレンズ群Ｇ１（レンズＬ１−１、Ｌ１−２から構成）とプリズムＰ３とレンズ群Ｇ２（レンズＬ２−１、Ｌ２−２、Ｌ２−３、Ｌ２−４、Ｌ２−５から構成）毎に、各面の偏心を測定した結果を示している。レンズ群Ｇ１、Ｇ２内における偏心は既存の測定機を用いて導出することが可能であり、また、プリズムＰ３に関しても、例えば図９で示した手法により偏心を導出することは可能である。導出された結果から、各面毎のローカル座標系の相対位置が分かることになり、図２１（ａ）〜（ｃ）に示すように、例えば、レンズ群Ｇ１では、レンズＬ１−１を構成する面のローカル座標系ＡＬ１−１−１、ＡＬ１−１−２と、レンズＬ１−２を構成する面のローカル座標系ＡＬ１−２−１、ＡＬ１−２−２の相対位置が分かる。また、例えば、プリズムＰ３では、プリズムＰ３を構成する面のローカル座標系ＡＰ３−１、ＡＰ３−２、ＡＰ３−３の相対位置が分かる。また、レンズ群Ｇ２では、レンズＬ２−１を構成する面のローカル座標系ＡＬ２−１−１、ＡＬ２−１−２と、レンズＬ２−２を構成する面のローカル座標系ＡＬ２−２−１、ＡＬ２−２−２と、レンズＬ２−３を構成する面のローカル座標系ＡＬ２−３−２と、レンズＬ２−４を構成する面のローカル座標系ＡＬ２−４−１、ＡＬ２−４−２と、レンズＬ２−５を構成する面のローカル座標系ＡＬ２−５−１、ＡＬ２−５−２の相対位置が分かる。
【００９５】
このとき、各面のローカル座標系の相対位置がレンズ群やプリズム毎に既知であるため、図２０に示すように、レンズＬ１−１、Ｌ２−５、プリズムＰ３の各被検面ＳＬ１−１、ＳＰ３−２、ＳＬ２−５のローカル座標系ＡＬ１−１−１、ＡＰ３−２、ＡＬ２−５−２との相対を組み合わせることで、レンズ群Ｇ１を構成する被検面ＳＬ１−１以外の面及びプリズムＰ３を構成するＳＰ３−２以外の面及びレンズ群Ｇ２を構成するＳＬ２−５以外の面を含む各ローカル座標系（ＡＬ１−１−１、ＡＬ１−１−２、ＡＬ１−２−１、ＡＬ１−２−２、ＡＰ３−１、ＡＰ３−２、ＡＰ３−３、ＡＬ２−１−１、ＡＬ２−１−２、ＡＬ２−２−１、ＡＬ２−２−２、ＡＬ２−３−２、ＡＬ２−４−１、ＡＬ２−４−２、ＡＬ２−５−１、ＡＬ２−５−２）の相対位置を求めることができる。これらの手法により、複数の素子における相対的な偏心量が分かるため、光学性能劣化の要因解明、修正等に反映させることが可能となる。また、本手法の結果を基に製造を実施することで、精度の高い光学系を作製することが可能となる。
【００９６】
なお、各被検面の形状を測定する際、被検面が略上を向くように配置させる必要がある場合には、図示しない回転機構を用いると示したが、測定時に基準ブロックＤと被検物の相対位置が変化しないで固定できればよいので、このような保持が可能であれば回転機構を用いなくてもよく、例えば、配置する姿勢を変えて固定してもよい。
【００９７】
ところで、図１や図２のような非接触３次元測定機の光プローブ２を走査して被検物Ｏの被検面１（Ｓ１、Ｓ２）の３次元面形状を測定するには、被検面１の全面を連続的に隈なく走査することは測定時間等の関係で困難であるので、通常は、測定面内のサンプリングされた飛び飛びの点で測定するのが実際である。その飛び飛びの点間の光プローブ２の移動に関しては、従来は、図２４（ｂ）に示すように、特定の位置で高さを測定した▲１▼の状態の光プローブ２をそのままの高さを保ったまま水平に移動して▲２▼の状態で次の位置の測定を行う。しかしながら、このような測定点間の光プローブ２の移動では、被検面１の勾配が急である場合には、光プローブ２を出す対物レンズ１５、２４が被検面１と接触して被検面１あるいは対物レンズ１５、２４を傷つけたり、被検面１との距離が遠くなりすぎ反射した光が再び対物レンズ１５、２４に取り込めずに測定ができなくなる場合があり得る。
【００９８】
そこで、本発明の測定方法においては、図２４（ａ）に示すように、特定の位置で高さを測定した▲１▼の状態の光プローブ２を所定の微小距離αだけ上方へ移動させて▲２▼の状態にし、その後、被検面１の設計形状から微小距離αだけ離れてその形状に沿うように移動させて次の測定位置に持ってきて▲３▼の状態にし、その状態で次の測定位置の高さの測定を行うようにする。ここで、▲２▼から▲３▼の状態への移動は、最短距離で移動させても、あるいは、設計値＋αの面に沿うように移動させてもよい。このような光プローブ２の移動方法をとることにより、対物レンズ１５、２４が被検面１と接触すること、及び、反射光が取り込めなくなることが何れも回避可能になる。なお、微小距離αは０以上で、被検面１からの反射光を非接触３次元測定機の対物レンズで取り込むことができる範囲において、任意に設定することができる。
【００９９】
次に、被検物Ｏの被検面上のローカル座標系で指定した座標位置に、マークｖを加工しておいたサンプルを使った測定方法について説明する。
【０１００】
図２５は、マークｖを加工しておいたサンプルの一例である。図２５に示すように、本発明の測定方法では、被検物Ｏの被検面上のローカル座標系で指定した座標位置にマークｖを加工しておいたサンプルを用いてもよい。この場合、マーク位置を、被検面の面形状測定時に同じ測定座標系で測定し、各マーク位置の測定値から、被検面のローカル座標系の位置を求めることができる。従って、ローカル座標系導出の際に、フィッティングや最適化等の解析が必要ないため、解析が容易となる。
【０１０１】
なお、マーク形状は、上記解析が可能であれば、形状や個数を限定するものではなく、例えば、形状は円柱、角柱、円錐、角錐、十字状、Ｌ字状、交線、点等でもよい。また、測定が高精度にできれば凸形状、凹形状のどちらであってもよい。また、マーク加工位置は、光学系の有効エリアＢ１、Ｂ２以外の領域に加工することが望ましい。マーク位置は、例えば、マーク形状が円柱の場合には、上面の外形の円の中心位置としてもよい。また、十字状の場合には、各直線の交点の位置としてもよい。
【０１０２】
なお、各マークは、被検面の面形状測定時に同じ測定座標系でマーク形状を測定できれば、必ずしも被検面と同一の面内にある必要はなく、例えば、被検物Ｏを構成する別の面にあってもよい。
【０１０３】
図２６は、マークｖを加工しておいたサンプルの別の一例である。図２６に示すように、発明の測定方法では、被検物Ｏの被検面上にローカル座標系で指定した座標位置にマークｖを加工しておいたサンプルを用いてもよい。この場合、マーク位置を、被検面の面形状測定時に同じ測定座標系で測定し、各マーク位置の測定値から、被検面のローカル座標系の位置を求めることができる。よって、被検面Ｓ１、Ｓ２上の各マークｖの測定結果より、各被検面を示すローカル座標系Ａ１，Ａ２の位置を求めておく。
【０１０４】
また、基準ブロックの形状は、図２７に示すように、底面が多面体からなる角柱であってもよい。この場合、被検物Ｏを保持する際に、各被検面Ｓ１、Ｓ２と、それぞれ基準ブロックＤを構成する少なくとも１つの面が略水平となるように保持する。ここで、各被検面Ｓ１、Ｓ２と、それぞれ略水平になった基準ブロックＤの面を基準面Ｈとする。
【０１０５】
図２７では、基準ブロックを構成する面Ｈ１、Ｈ２が、それぞれ各被検面Ｓ１、Ｓ２と略水平に保持されているため、各被検面の基準面Ｈとして使用する。このような構成にすることにより、被検物Ｏの形状にあわせて基準ブロックを製作する必要がなく、汎用的に同一の基準ブロックを使用することができる。なお、図２７では、基準ブロックの形状を、８角柱としたが、上記保持が可能であれば、これに限定するものではない。
【０１０６】
次に、図２８〜図３１を参照にして、枠Ｗに光学素子( 例えばプリズム) が複数配置された被検物Ｏの、各光学素子の偏心量を、被検物Ｏを保持する枠Ｗに固定した基準となる素子( 基準ブロック) Ｄを用いて算出する第２の方法について説明する。このとき、被検物Ｏは、ある方向において、各光学素子を構成する少なくとも１面に加工しておいたマーク位置を測定できるようにする。
【０１０７】
なお、マーク形状は、上記測定が可能であれば、形状や個数を限定するものではなく、例えば、形状は円柱、角柱、円錐、角錐、十字状、Ｌ字状、交線、点等がある。また、測定が高精度にできれば凸形状、凹形状のどちらであってもよい。また、マーク加工位置は光学系の有効エリアＢ１−１、Ｂ２−１以外の領域に加工することが望ましい。
【０１０８】
図２８〜図３１では、プリズムＰ１とプリズムＰ２が枠Ｗ内に配置され、プリズムＰ１を構成する面Ｓ１−１とプリズムＰ２を構成する面Ｓ２−１上にそれぞれ加工されたマークｖが外側に出ており、光プローブを照射することで各マーク形状の測定が可能となっており、その結果マークｖの位置を求めることが可能である。
【０１０９】
図２８は、プリズムＰ１の被検面Ｓ１−１と基準ブロックＤの１つの基準面Ｈに対して光プローブを走査することにより、被検面のＳ１−１上にあるマーク形状、基準面Ｈのエッジe の形状を測定する様子を示す斜視図、図２９は被検面Ｓ１−１の上面図、図３０はプリズムＰ２の被検面Ｓ２−１と基準ブロックＤの他の基準面Ｈに対して光プローブを走査することにより、被検面のＳ２−１上にあるマーク形状、基準面Ｈのエッジｅの形状を測定する様子を示す斜視図、図３１は被検面Ｓ２−１の上面図である。
【０１１０】
被検物Ｏは基準ブロックＨと保持具Ｃで一体に固定して保持されており、基準ブロックＤの各基準面Ｈとそれに対応した各被検面Ｓ１−１、Ｓ２−１上にあるマーク形状をそれぞれ非接触３次元測定機を用いて同じ測定座標系で測定できるようにして、マーク形状の測定値よりマーク位置を求める。
【０１１１】
基準ブロックの形状は、被検物ＯのプリズムＰ１、Ｐ２の各被検面Ｓ１−１、Ｓ２−１の測定と同時に少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点の位置を測定できるように構成することが望ましい。このとき、測定に用いた少なくとも２つのエッジｅ若しくは１つのエッジｅと座標が既知である点により、基準ブロックＤの各基準面にＨの位置を定義できるようなものとする必要がある。さらに少なくとも各測定面Ｈと同時に測定するエッジｅの角度や長さは、予め高精度に測定しておき、これらの既知のデータから測定値で定義される各基準面Ｈの相対的な位置が分かるようにしておく。
【０１１２】
次に測定方法について説明する。図１１〜図１４で示した例と同様の手法により、各被検面Ｓ１−１、Ｓ２−１毎に基準ブロックＤのエッジｅの形状と、被検物ＯであるプリズムＰ１、Ｐ２の各被検面Ｓ１−１、Ｓ２−１上にある各マークｖを、同じ非接触３次元測定機を用いて同じ測定座標系で測定する。このとき、例えば測定機の関係で、光プローブの入射する方向に対して被検面を略垂直になるように、被検面Ｓ１−１、Ｓ２−１を上に向ける必要がある場合、図２８〜図３１に示すように、各面の測定毎に被検物Ｏを回転して保持する必要がある。そのため、図示しない回転機構を用いて、被検物Ｏと基準ブロックＤを所望の方向に回転させて固定することになる。このとき、被検物Ｏと基準ブロックＤは、回転によっても相対的な位置関係が変化しないようにする。
【０１１３】
なお、基準面Ｈの位置の導出には、図９で示した例と同様の手法を用いることができ、これにより、基準ブロックＤの各基準面Ｈの位置を求めることが可能となる。さらに、同じ測定座標系で被検面Ｓ１−１、Ｓ２−１上にある各マークｖの位置を測定することにより、被検面Ｈと被検面Ｓ１−１、Ｓ２−１上にあるマークｖの相対的な位置を各被検面Ｓ１−１、Ｓ２−１毎に把握することが可能となる。
【０１１４】
このマークｖについては、例えば図２６で示した同様の手法を用いることで、すなわち、各面の形状測定とともに、同じ測定座標系を用いてマーク位置を測定し、同様の解析をすることで、各面の面形状とともに、マークの相対位置を算出することが可能となる。したがって、被検面の面形状とマーク位置の相対位置を導出することが可能である。
【０１１５】
さらに、基準ブロックＤを構成する基準面Ｈの既知のデータに対応させて、各基準面Ｈの測定値により求められた各面の位置を変換し、各面の相対的な位置関係を合わせる。このとき、シフトや回転といった座標変換を行う必要がある。
【０１１６】
この解析により得られた座標変換を、各被検面Ｓ１−１、Ｓ２−１上にある各マークｖにも実施することで、各被検面Ｓ１−１、Ｓ２−１上にあるマークの相対的な位置を把握することが可能となる。
【０１１７】
また、枠Ｗに組み込む前に、各プリズムの各被検面のマーク位置とローカル座標系の相対位置を、例えば図２６で示した手法により求めておけば、各プリズム毎にプリズムを構成する各面のマークとローカル座標系の相対位置が既知となる。
【０１１８】
なお、マークが、被検面上のローカル座標系で指定した座標位置に加工してあれば、各マーク位置の測定結果より、容易にローカル座標系の位置を算出することが可能となる。そこで、例えば図２６で示した手法により、各プリズム毎にプリズムを構成する各面の複数のマークｖとローカル座標系の相対位置が分かる。
【０１１９】
ただし、各マークの、被検面のローカル座標系における座標位置が分からない場合でも、解析は可能である。この場合、マークの位置を測定する際に、同じ測定座標系を用いて、被検面の形状を測定するようにする。面形状データは、例えば座標変換によって、それぞれ各設計値形状に重ね合わせた際の残差が最小となるような最適化の解析を行うことで、被検面を表すローカル座標系の位置を求めることが可能となるので、各面毎にローカル座標系とマークｖの相対的な位置が分かる。そこで、例えば図２６で示した手法により、各プリズム毎にプリズムを構成する各面の面形状とローカル座標系とともに、各マークｖの相対位置が分かる。この場合、マークｖは、被検面と同一の面上になくても、同一の測定座標系で、マーク位置を測定できれば、同様の解析は可能である。
【０１２０】
そこで、これらの結果を図３０で示した各プリズムＰ１、Ｐ２の被検面Ｓ１−１、Ｓ２−１上にある同一の各マークの位置に対応させることで、図３１に示すように、プリズムＰ１を構成するＳ１−１以外の面（Ｓ１−２、Ｓ１−３）及びプリズムＰ２を構成するＳ２−１以外の面（Ｓ２−２、Ｓ２−３）の各面のローカル座標系（Ａ１−２、Ａ１−３、Ａ２−２、Ａ２−３）の相対位置も求めることができる。
【０１２１】
上記解析には、例えば対応する各被検面毎に、図３０で測定したマーク位置と、枠Ｗ組み込み前に測定した各プリズムでの対応する各マークを座標変換して、各マーク位置間の距離の総和が最小になるように座標変換パラメータを算出するような最適化の解析を行ってもよい。さらに、この解析によって導出した座標変換を、マーク位置と相対的な位置関係が既知である、被検面のローカル座標系や他の被検面のローカル座標系にも実施すればよい。
【０１２２】
これらの手法により、複数の素子における相対的な偏心量が分かるため、光学性能劣化の要因解明、修正等に反映させることが可能となる。また、本手法の結果を基に製造を実施することで精度の高い光学系を作製することが可能となる。
【０１２３】
なお、本手法では、枠Ｗに組み込み前後で被検面のマークの測定を行う必要があるが、マーク位置を算出する際にフィッティングや最適化等の解析を必要としないため、枠Ｗに組み込み前後での測定誤差の影響を少なくすることが可能となる。
【０１２４】
また、上記解析では、各プリズム毎に、マーク付の被検面を１面ずつ用いた場合を示したが、これに限定するものではなく、例えば１つのプリズムについて、複数の被検面上にあるマークを用いて解析を行ってもよい。この場合、マーク位置の対応の情報が増加する為、算出される複数の素子における相対的な偏心量の精度が向上する。
【０１２５】
以上の本発明の３次元座標測定方法は、例えば次のように構成することができる。
【０１２６】
〔１〕 被検物上のエッジ近傍にそのエッジに略沿う曲線を設定し、その曲線と略直交する方向に走査して、エッジに基づいて定められた位置の座標値を検出することを特徴とする３次元座標測定方法。
【０１２７】
〔２〕 前記曲線を多項式で定義することを特徴とする上記１記載の３次元座標測定方法。
【０１２８】
〔３〕 前記曲線が、エッジ近傍の複数の点を選択し、その複数の点をなだらかに結んで定義した曲線であることを特徴とする上記１記載の３次元座標測定方法。
【０１２９】
〔４〕 前記複数の点をなだらかに結ぶ際にスプライン演算を用いることを特徴とする上記３記載の３次元座標測定方法。
【０１３０】
〔５〕 被検物を構成する面の中、少なくとも１面が自由曲面からなることを特徴とする上記１記載の３次元座標測定方法。
【０１３１】
〔６〕 被検物が、構成面の中少なくとも１面が自由曲面からなる光学素子であることを特徴とする上記５記載の３次元座標測定方法。
【０１３２】
〔７〕 被検物が、構成面の中少なくとも１面が自由曲面からなる光学素子を製作するための金型であることを特徴とする上記５記載の３次元座標測定方法。
【０１３３】
〔８〕 複数の面から構成される被検物に対して、各面毎に３次元面形状とエッジ形状とをそれぞれ測定し、共通のエッジの測定値を対応させることにより前記複数の面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１３４】
〔９〕 異なる面毎に測定した共通のエッジの測定値が略同一の曲線上に乗るように対応付けをすることで、前記複数の面の相対的な位置関係を求めることを特徴とする上記８記載の３次元座標測定方法。
【０１３５】
〔１０〕 各面毎に測定した面形状測定値と面の設計値とを当てはめることで各面毎のローカル座標系を求め、１つの面で測定した共通のエッジの測定値と他の面で測定した共通のエッジの測定値とを対応させる座標変換を前記ローカル座標系にも行うことで、複数の面におけるローカル座標系の相対的な位置関係を求めることを特徴とする上記９記載の３次元座標測定方法。
【０１３６】
〔１０−１〕 各面毎に測定した面形状測定値と、座標変換した各被検面の設計形状を重ね合わせた際に、残差が最小となるような座標変換パラメータを求める最適化によって、各面毎のローカル座標系を求め、１つの面で測定した共通のエッジの測定値と他の面で測定した共通のエッジの測定値とを対応させる座標変換を前記ローカル座標系にも行うことで、複数の面におけるローカル座標系の相対的な位置関係を求めることを特徴とする上記９記載の３次元座標測定方法。
【０１３７】
〔１１〕 被検物を構成する面の中、少なくとも１面が自由曲面からなることを特徴とする上記９記載の３次元座標測定方法。
【０１３８】
〔１２〕 被検物が、構成面の中少なくとも１面が自由曲面からなる光学素子であることを特徴とする上記１１記載の３次元座標測定方法。
【０１３９】
〔１３〕 被検物が、構成面の中少なくとも１面が自由曲面からなるを製作するための金型であることを特徴とする上記１１記載の３次元座標測定方法。
【０１４０】
〔１４〕 複数の面から構成される被検物と、基準面の相対的な位置関係が既知な基準素子とを相対的に固定し、被検物の各被検面の３次元面形状と基準素子の形状とを同じ測定座標系でそれぞれ測定し、基準素子のエッジ形状の測定値から求まる面形状データに対応させて基準素子の各基準面の座標を変換し、その同じ座標変換を各被検面の３次元面形状測定値にも行うことにより、被検物の複数の被検面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４１】
〔１４−１〕 複数の面から構成される被検物と、基準面の相対的な位置関係が既知な基準素子と被検物とを相対的に固定し、被検物の各被検面の３次元面形状と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面のマークの位置にも行うことにより、被検物の複数の被検面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４２】
〔１４−２〕 複数の面から構成される被検物において、基準面の相対的な位置関係が既知な基準素子と被検物とを相対的に固定し、被検物の各被検面上にあるマークの位置と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面のマークの位置にも行うことにより、被検物の複数の被検面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４３】
〔１４−３〕 複数の面から構成される被検物において、基準面の相対的な位置関係が既知な基準素子と被検物とを相対的に固定し、被検物の各被検面上にあるマークの位置と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子のエッジ形状の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面のマークの位置にも行うことにより、被検物の複数の被検面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４４】
〔１５〕 複数の面から構成される素子を複数個用いて構成する被検物において、基準面の相対的な位置関係が既知な基準素子と被検物を相対的に固定し、被検物の各素子の少なくとも１面の３次元形状と基準素子とを同じ座標系でそれぞれ測定し、基準素子のエッジ形状の測定値から求まる面形状データに対応させて基準素子の各基準面の座標を変換し、その同じ座標変換を各被検面の３次元面形状測定値にも行うことにより、各素子を構成する少なくとも１面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４５】
〔１５−１〕 複数の面から構成される素子を複数個用いて構成する被検物において、基準面の相対的な位置関係が既知な基準素子と被検物を相対的に固定し、被検物の各素子の少なくとも１面の３次元形状と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面の３次元形状測定値にも行うことにより、各素子を構成する少なくとも１つの面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４６】
〔１５−２〕 複数の面から構成される素子を複数個用いて構成する被検物において、基準面の相対的な位置関係が既知な基準素子と被検物を相対的に固定し、被検物の各素子のそれぞれ少なくとも１つの被検面上にあるマークの位置と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面のマークの位置にも行うことにより、各素子を構成するそれぞれ少なくとも１つの面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４７】
〔１５−３〕 複数の面から構成される素子を複数個用いて構成する被検物において、基準面の相対的な位置関係が既知な基準素子と被検物を相対的に固定し、被検物の各素子のそれぞれ少なくとも１つの被検面上にあるマークの位置と基準素子とを同じ測定座標系でそれぞれ測定し、基準素子のエッジ形状の測定値から求まる面形状データに対応させて各基準素子の基準面の座標を変換し、その同じ座標変換を各被検面のマークの位置にも行うことにより、各素子を構成するそれぞれ少なくとも１つの面の相対的な位置関係を求めることを特徴とする３次元座標測定方法。
【０１４８】
〔１６〕 基準素子の少なくとも２つのエッジ、若しくは、１つのエッジと座標が既知である点の位置とを測定することにより基準素子の形状を求めることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１４９】
〔１７〕 基準素子を構成する各基準面のなす角度が既知であることを特徴とする上記１６記載の３次元座標測定方法。
【０１５０】
〔１８〕 基準面の測定するエッジは、各基準面からなる稜線を含むことを特徴とする上記１７記載の３次元座標測定方法。
【０１５１】
〔１９〕 基準面の少なくとも２つのエッジ、若しくは、１つのエッジと座標が既知である点の測定結果から、各基準面からなる稜線の位置と基準面を表す平面の位置を導出することを特徴とする上記１８記載の３次元座標測定方法。
【０１５２】
〔２０〕 同一の稜線を含む各基準面の測定結果からそれぞれ導出した稜線の位置が空間的に一致するように座標の変換を行い、これにより空間的に一致した稜線を回転軸として、稜線を含む各平面を、基準素子の既知の値と対応するように、座標の変換を行うパラメータを算出することを特徴とする上記１９記載の３次元座標測定方法。
【０１５３】
〔２１〕 前記座標の変換を、各基準面と同じ測定座標系で測定した各被検面の測定結果に対して実施することを特徴とする上記２０記載の３次元座標測定方法。
【０１５４】
〔２２〕 前記座標の変換を、各基準面と同じ測定座標系で測定した各被検面の測定結果から求めたローカル座標系に対して実施することを特徴とする上記２０記載の３次元座標測定方法。
【０１５５】
〔２３〕 各面毎に測定した面形状測定値と面の設計値とを当てはめることで各面毎のローカル座標系を求め、求めたローカル座標系に対して、各被検面の３次元面形状測定値に対して行う座標変換を行うことで、被検物の複数の被検面におけるローカル座標系の相対的な位置関係を求めることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１５６】
〔２３−１〕 各被検面上のローカル座標系で指定した座標位置にマークを加工しておいたサンプルを用いて、各面毎に測定したマーク位置から、ローカル座標系を求め、求めたローカル座標系に対して、各基準面に対して行う座標変換を行うことで、被検物の複数の被検面におけるローカル座標系の相対的な位置関係を求めることを特徴とする上記１４又は１４−２又は１５又は１５−２記載の３次元座標測定方法。
【０１５７】
〔２３−２〕 各面毎に測定した面形状測定値と、座標変換した各被検面の設計形状を重ね合わせた際に、残差が最小となるような座標変換パラメータを求める最適化によって、各面毎のローカル座標系を求め、求めたローカル座標系に対して、各基準面に対して行う座標変換を行うことで、被検物の複数の被検面におけるローカル座標系の相対的な位置関係を求めることを特徴とする上記１４又は１４−２又は１５又は１５−２記載の３次元座標測定方法。
【０１５８】
〔２４〕 被検物を構成する面の中、少なくとも１面が自由曲面からなることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１５９】
〔２５〕 被検物が、構成面の中少なくとも１面が自由曲面からなる光学素子であることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６０】
〔２６〕 被検物が、構成面の中少なくとも１面が自由曲面からなるを製作するための金型であることを特徴とする上記２４記載の３次元座標測定方法。
【０１６１】
〔２７〕 基準素子を構成する基準面が平面により構成されていることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６２】
〔２７−１〕 基準素子は、底面が多角形からなる角柱であることを特徴とする上記１４又は１４−３又は１５又は１５−２記載の３次元座標測定方法。
【０１６３】
〔２８〕 基準素子の形状は、被検面を平面で近似した構成の多面体であることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６４】
〔２９〕 基準素子の形状は、被検面のローカル座標系で定義される平面で構成される多面体であることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６５】
〔３０〕 基準素子はガラスでできていることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６６】
〔３１〕 基準素子は金属でできていることを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６７】
〔３２〕 被検物と基準素子は、回転機構を有する治具に保持することを特徴とする上記１４、１４−３、１５、１５−３のいずれかに記載の３次元座標測定方法。
【０１６８】
〔３３〕 非接触光プローブを用いて測定を行い、その際に特定の測定点から次の測定点へ非接触光プローブを移動させるのに、特定の測定点での非接触光プローブを所定の微小距離だけ被検面から離れる方向へ移動させ、次いで被検面の設計値からその微小距離だけ離れてその設計値に沿うように移動させて次の測定点へ移動させ、その状態で次の測定点の測定を行うことを特徴とする上記１から３２の何れか１項記載の３次元座標測定方法。
【０１６９】
〔３４〕 被検物を構成する各素子を構成する各面の相対的な位置関係が既知であることを特徴とする上記１５又は１５−３に記載の３次元座標測定方法。
【０１７０】
〔３５〕 複数の面から構成される素子を複数個用いて構成された被検物は、枠で保持されていることを特徴とする上記１６記載の３次元座標測定方法。
【０１７１】
〔３６〕 被検物を保持する枠は、被検面に光が照射できる構成にしたことを特徴とする上記１６記載の３次元座標測定方法。
【０１７２】
〔３７〕 被検物が回転対称な光学面を有することを特徴とする記１５又は１５−３に記載の３次元座標測定方法。
【０１７３】
〔３８〕 被検物が折り曲げられた光軸を有する特徴とする記１５又は１５−３に記載の３次元座標測定方法。
【０１７４】
〔３９〕 非接触光プローブで被検面の３次元座標を測定する３次元座標測定方法において、特定の測定点から次の測定点へ非接触光プローブを移動させる際に、特定の測定点での非接触光プローブを所定の微小距離だけ被検面から離れる方向へ移動させ、次いで被検面の設計値からその微小距離だけ離れてその設計値に沿うように移動させて次の測定点へ移動させ、その状態で次の測定点の測定を行うことを特徴とする３次元座標測定方法。
【０１７５】
〔４０〕 上記１から３９の何れか１項記載の３次元座標測定方法を用いた測定機。
【０１７６】
〔４１〕 上記１から３９の何れか１項記載の３次元座標測定方法で測定されたもの。
【０１７７】
【発明の効果】
以上の説明から明らかなように、本発明の３次元座標測定方法によると、外形に対する各被検面の設計座標系（ローカル座標系）の相対位置を正確に求めることができ、また、複数の被検面におけるローカル座標系の相対的な位置関係を明確にすることが可能となる。そのため、複数の被検面における相対的な偏心量が分かることになり、光学性能劣化の要因解明、修正等に反映させることが可能となる。
【図面の簡単な説明】
【図１】本発明において使用可能なオートフォーカスタイプの非接触３次元測定機の構成を概略的に示す図である。
【図２】本発明において使用可能な共焦点顕微鏡タイプの非接触３次元測定機の構成を概略的に示す図である。
【図３】円形状のエッジ部を測定する方法を説明するための図である。
【図４】本発明に基づいてエッジの形状を測定する方法を説明するための図である。
【図５】エッジの形状を測定する際の走査開始点を示す曲線の設定方法を説明するための図である。
【図６】本発明に基づいて外形に対する各面のローカル座標系の相対位置を求める方法を説明するための図である。
【図７】本発明に基づいて被検物の１つのエッジを共有する複数の被検面の偏心量を算出する方法を説明するための図である。
【図８】図７の方法で相対的な位置関係が明確になった複数の面のローカル座標系を示す図である。
【図９】本発明に基づいて被検物に固定した基準ブロックを用いて複数の被検面の偏心量を算出する方法を説明するための図である。
【図１０】図９の方法で相対的な位置関係が明確になった複数の面のローカル座標系を示す図である。
【図１１】本発明に基づいて被検物に固定した基準ブロックを用いて複数の素子の各素子を構成する被検面の偏心量を導出する方法を説明するための図中の、１つの素子の被検面と基準ブロックの１つの基準面に対して光プローブを走査することによりその被検面の形状、基準面のエッジの形状を測定する様子を示す斜視図である。
【図１２】図１１の被検面の上面図である。
【図１３】図１１の別の素子の被検面と基準ブロックの他の基準面に対して光プローブを走査することによりその被検面の形状、基準面のエッジの形状を測定する様子を示す斜視図である。
【図１４】図１３の被検面の上面図である。
【図１５】図１１〜図１４の方法で相対的な位置関係が明確になった複数の素子の各素子を構成する面のローカル座標系の相対位置を示す図である。
【図１６】図１１〜図１４の方法で、予め各素子を構成する面の偏心を導出しておいた素子からなる光学系を被検物として用いることにより、相対的な位置関係が明確になった複数の素子を構成する各面のローカル座標系の相対位置を示す図である。
【図１７】本発明に基づいて被検物に固定した基準ブロックを用いて非球面レンズを構成する被検面の偏心量を導出する方法を説明するための図である。
【図１８】図１７の方法で被検面が球面であった場合に、被検面のローカル座標系を導出するために加工したマークや被検物の外形を示す図である。
【図１９】２つのレンズ群とプリズムからなる被検物を説明するための図である。
【図２０】本発明に基づいて被検物に固定した基準ブロックを用いて図１９に示した２つのレンズ群とプリズムからなる被検物を構成する被検面の偏心量を導出する方法を説明するための図である。
【図２１】図２０に示した２つのレンズ群とプリズムについてそれぞれ各素子（レンズ、プリズム）を構成する各面のローカル座標系を説明するための図である。
【図２２】図２０の方法で相対的な位置関係が明確になった複数の群とプリズムの各素子を構成する面のローカル座標系の相対位置を示す図である。
【図２３】図２０の方法で予め各素子を構成する面の偏心を導出しておいた素子からなる光学系を被検物として用いることにより、相対的な位置関係が明確になった複数の素子を構成する各面のローカル座標系の相対位置を示す図である。
【図２４】本発明における光プローブの移動方法を説明するための図である。
【図２５】マークｖを加工しておいたサンプルの一例を用いて、マークの位置に対する各面のローカル座標系の相対位置を求める方法を説明する為の図である。
【図２６】マークｖを加工しておいたサンプルの別の一例を用いて、被検物に固定した基準ブロックを用いて複数の被検面の偏心量を算出する方法を説明する為の図である。
【図２７】本発明に基づいて被検物に固定した底面が多角形からなる角柱の基準ブロックを用いて複数の被検面の偏心量を算出する方法を説明する為の図である。
【図２８】本発明に基づいて被検物に固定した基準ブロックを用いて複数の素子の各素子を構成する被検面の偏心量を導出する方法を説明するための図中の、１つの素子の被検面と基準ブロックの１つの基準面に対して光プローブを走査することによりその被検面上にあるマークの位置、基準面の形状を測定する様子を示す斜視図である。
【図２９】図２８の被検面の上面図である。
【図３０】図２８の別の素子の被検面と基準ブロックの他の被検面に対して光プローブを走査することによりその被検面上にあるマークの位置、基準面の形状を測定する様子を示す斜視図である。
【図３１】図３０の被検面の上面図である。
【符号の説明】
Ｏ…被検物
Ｓ１、Ｓ２、Ｓ１−１、Ｓ２−１、Ｓ１−２、Ｓ１−３、Ｓ２−２、Ｓ２−３、Ｓ３−１、Ｓ３−２、ＳＬ１−１、ＳＰ３−２、ＳＬ２−5 …被検面
Ｐ１、Ｐ２、Ｐ３…プリズム
Ｌ１−１、Ｌ１−２、Ｌ２−１、Ｌ２−２、Ｌ２−３、Ｌ２−４、Ｌ２−５…レンズ
Ｇ１、Ｇ２…レンズ群
Ａ１、Ａ２、Ａ１−１、Ａ２−１、Ａ１−２、Ａ１−３、Ａ２−２、Ａ２−３、Ａ３−１、Ａ３−２、ＡＬ１−１−１、ＡＬ１−１−２、ＡＬ１−２−１、ＡＬ１−２−２、ＡＰ３−１、ＡＰ３−２、ＡＰ３−３、ＡＬ２−１−１、ＡＬ２−１−２、ＡＬ２−２−１、ＡＬ２−２−２、ＡＬ２−３−２、ＡＬ２−４−１、ＡＬ２−４−２、ＡＬ２−５−１、ＡＬ２−５−２…ローカル座標系
Ｗ…枠
Ｍ…基準マーク
Ｅ、Ｅ１、Ｅ２、ｅ…エッジ
Ｌ、Ｌ１、Ｌ２…曲線
Ｂ、Ｂ１、Ｂ２、Ｂ１−１、Ｂ２−１…有効エリア
Ｄ…基準ブロック
Ｈ、Ｈ１、Ｈ２…基準面
Ｃ…保持具
Ｆ…円
ｖ…マーク
１…被測定面
２…光プローブ
１１…レーザ
１２、１３、１４…ミラー
１５…対物レンズ
１６…移動機構
１７…光位置検出装置
１８…ＸＹステージ
２１…光源
２２…第１ピンホール
２３…ハーフミラー
２４…対物レンズ
２５…第２ピンホール
２６…光検出器
２７…移動機構
２８…ＸＹステージ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a three-dimensional coordinate measuring method, and in particular, a surface shape based on an edge of an optical element or a mold used when manufacturing them by plastic molding, glass molding or the like, and a relative positional relationship between a plurality of surfaces. The present invention relates to a three-dimensional coordinate measurement method for obtaining the above.
[0002]
[Prior art]
Since the surface shape of an optical element such as a lens or prism greatly affects the performance of the optical system, measurement of the absolute surface shape is an important issue in quality control in the manufacturing process of the element. As a method for measuring the surface shape of the optical element, an interferometer has been conventionally used. However, the interferometer is a relative comparison with the reference surface, and the absolute shape cannot be measured.
[0003]
As a method for measuring the absolute shape, a stylus type shape measuring machine is commercially available, but mainly measures the shape of a cross section perpendicular to the optical axis, and measures the surface shape in two dimensions. Therefore, the asymmetric surface shape of the optical element cannot be measured correctly. There was also a similar problem with the mold.
[0004]
On the other hand, as a device capable of measuring the three-dimensional coordinates of the sample surface with high accuracy, there is a three-dimensional measuring machine. However, since there is generally no absolute coordinate system of the device itself, there is a problem that the absolute shape cannot be measured.
[0005]
In such a situation, the applicant of the present invention, in Japanese Patent Application No. 2000-383734, Japanese Patent Application No. 2001-296030, and Japanese Patent Application No. 2001-299727, simultaneously measures an edge or a mark with a non-contact optical probe, and the measurement value is determined by the edge or the like. A method for obtaining the absolute shape of an optical element by converting it into coordinates has been proposed.
[0006]
[Problems to be solved by the invention]
However, in Japanese Patent Application No. 2000-383734 and the like, a method for obtaining a relative positional relationship between coordinate systems expressing the surface shape of an optical element or a mold has not been studied in detail.
[0007]
The present invention has been made in view of such a situation in the prior art, and its purpose is to form a surface shape based on an optical element as a test object or an edge of the mold, and a relative positional relationship between a plurality of surfaces. It is to provide a three-dimensional coordinate measuring method for obtaining the above.
[0008]
[Means for Solving the Problems]
In the first three-dimensional coordinate measuring method of the present invention that achieves the above object, a curve that substantially follows the edge is set in the vicinity of the edge on the test object, scanned in a direction substantially perpendicular to the curve, It is a method characterized by detecting the coordinate value of the position determined based on it.
[0009]
In the second three-dimensional coordinate measuring method of the present invention, a three-dimensional surface shape and an edge shape are measured for each surface of a test object composed of a plurality of surfaces, and a common edge measurement value is measured. To obtain the relative positional relationship between the plurality of surfaces.
[0010]
According to the third three-dimensional coordinate measurement method of the present invention, a test object composed of a plurality of surfaces and a reference element whose relative positional relationship is known are relatively fixed, and the test object is measured. The three-dimensional surface shape of each test surface and the shape of the reference element are measured in the same measurement coordinate system, and the coordinates of each reference surface of the reference element corresponding to the surface shape data obtained from the measured value of the edge shape of the reference element And calculating the relative positional relationship of the plurality of test surfaces of the test object by performing the same coordinate conversion on the three-dimensional surface shape measurement value of each test surface. is there.
[0011]
Here, “the same measurement coordinate system” means that a common coordinate system is used, and for example, even when different coordinate systems are used for measurement of the test surface and the reference surface, they are used for each measurement. If the relative position of the coordinate system is known, the position of two coordinate systems can be expressed by one coordinate system, which also means such a case.
[0012]
A fourth three-dimensional coordinate measurement method of the present invention is a three-dimensional coordinate measurement method for measuring a three-dimensional coordinate of a surface to be measured with a non-contact optical probe. When moving, the non-contact optical probe at a specific measurement point is moved away from the test surface by a predetermined minute distance, and then separated from the design value of the test surface by the minute distance and along the design value. And moving to the next measurement point, and measuring the next measurement point in that state.
[0013]
According to a fifth three-dimensional coordinate measuring method of the present invention, in a test object configured by using a plurality of elements composed of a plurality of surfaces, a reference element whose test object has a known relative positional relationship between the reference surface and the test object Are fixed relative to each other, and the three-dimensional shape of at least one surface of each element of the test object and the reference element are measured in the same coordinate system to correspond to the surface shape data obtained from the measured value of the edge shape of the reference element. By converting the coordinates of each reference surface of the reference element and performing the same coordinate conversion on the three-dimensional surface shape measurement value of each test surface, the relative positional relationship of at least one surface constituting each element can be obtained. It is a method characterized by obtaining.
[0014]
In the present invention, it is possible to accurately obtain the relative position of the design coordinate system (local coordinate system) of each test surface with respect to the outer shape, and to clarify the relative positional relationship of the local coordinate system on a plurality of test surfaces. It becomes possible to. Therefore, the relative amount of eccentricity on a plurality of test surfaces can be known, and can be reflected in elucidation of the cause of optical performance degradation, correction, and the like.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a three-dimensional coordinate measuring method of the present invention will be described based on examples with reference to the drawings.
[0016]
As a three-dimensional measuring machine used in the measuring method of the present invention, an autofocus type non-contact three-dimensional measuring machine that measures the position of an object with a non-contact optical probe, a confocal microscope type non-contact three-dimensional measuring machine, or the like There is. First, these principles will be described.
[0017]
FIG. 1 is a diagram schematically showing a configuration of an autofocus type non-contact three-dimensional measuring machine (for example, Japanese Patent Application Laid-Open No. 2000-146532). Laser light emitted from a laser 11 passes through mirrors 13 and 12. The light enters the objective lens 15, proceeds toward the focal plane of the objective lens 15 at the center of the optical axis, is incident on the surface to be measured 1 of the test object O, is reflected, passes through the objective lens 15 again, and is reflected by the mirrors 12, 13. , 14, and forms an image on the optical position detection device 17. When the focus of the objective lens 15 is not in alignment with the surface 1 to be measured, the imaging position of the optical position detection device 17 changes. Therefore, the optical position detection device 17 catches this position change, and the objective lens 15 is moved by the moving mechanism 16. The height of the surface to be measured 1 in the Z-axis direction can be measured based on the position of the objective lens 15 when it is in focus, and the objective of the XY stage 18 on which the object to be tested O is placed is measured. The position of the XY plane can be determined by the position of the lens 15 with respect to the optical axis. In this way, the three-dimensional shape of the measurement surface 1 of the test object O can be measured.
[0018]
FIG. 2 is a diagram schematically showing the configuration of a confocal microscope type non-contact three-dimensional measuring machine. Light emitted from the light source 21 irradiates the first pinhole 22 and passes through the first pinhole 22. The light is reflected by the half mirror 23, enters the objective lens 24, travels toward the image point position of the objective lens 24 at the center of the optical axis, enters the measurement surface 1 of the test object O, reflects, and again It passes through the objective lens 24, passes through the half mirror 23, and is arranged at a position conjugate with the first pinhole 22. First It passes through the 2-pin hole 25 and enters the photodetector 26. When the surface 1 to be measured is not in a position conjugate with the first pinhole 22, the light emitted from the first pinhole 22 cannot pass through the second pinhole 25, so the photodetector 26 does not detect the light. The entire system is moved in the optical axis direction by the moving mechanism 27, and the height in the Z-axis direction of the measurement surface 1 can be measured by the position when the light detector 26 detects the light. The position of the XY plane can be determined by the position of the XY stage 28 on which O is placed with respect to the optical axis of the objective lens 24. In this way, the three-dimensional shape of the measurement surface 1 of the test object O can be measured.
[0019]
The light beam for position detection of the non-contact three-dimensional measuring machine as described above is called an optical probe. By using the non-contact optical probe to scan the measurement surface 1 of the test object O, measurement data representing the surface shape is obtained. Further, at the edge portion of the test object O, when the optical probe is scanned, the height value greatly changes before and after the sampling point. For example, a threshold value based on the height at the scanning start point is set. It is possible to detect the position of the edge portion by defining the point where the height value is larger than the threshold and the height value is changed as the edge portion.
[0020]
Therefore, for example, in order to efficiently measure the circular edge portion E, it is necessary to input the approximate center coordinates of the circle and the approximate radius as shown in FIG. Here, the center coordinates may be approximately the center of the circle to be measured. Then, a circle F for starting scanning of the optical probe for measuring the edge E is defined by the inputted center coordinates and radius. However, depending on the shape of the test surface, as shown in FIG. 3 (a), when scanning in the direction outside the center (in the case of a disk), the value of the input radius is determined from the radius of the circle of the test object. Is slightly smaller so that the scanning start point is inside the circle of the test object.
[0021]
Also, as shown in FIG. 3B, when scanning from the outside toward the center (in the case of a circular hole), the value of the input radius is made slightly larger than the radius of the circle of the test object, The scanning start point is set outside the circle of the test object.
[0022]
In the measurement, a plurality of scanning start points are set on the circle F, and scanning is performed from these in the direction of the edge E (in the outward direction in the case of FIG. 1 (a), inward direction in the case of FIG. 1 (b)). The coordinates of the edge E are obtained by comparing threshold values and the like. By fitting the coordinates of the edge E obtained by scanning of these optical probes to a circle using, for example, the least square method, it becomes possible to calculate the center coordinates, roundness, etc. of the test object.
[0023]
However, when the shape of the edge of the test surface is a curve that is not a circle or a straight line, the same measurement is difficult. For example, in the case of an optical element such as a free-form surface, a cylindrical surface, an anamorphic surface, an aspherical surface, or a prism formed by decentering a spherical surface, the edge of the surface constituting the prism can be represented only by a straight line or a circle. It cannot be a curve that is expressed in polynomial form.
[0024]
Below, the example which measures the shape of the edge which consists of arbitrary curves with respect to such a test object is demonstrated.
[0025]
A curve L along the shape of the curve indicating the edge E of the test object (prism) O as shown in FIGS. 4A and 4B is defined. 4A is a perspective view showing a state in which the shape of the surface and the shape of the edge are measured by scanning the optical probe 2 with respect to the test object O, and FIG. It is a top view of test surface S1. Here, the shape of the curve L may be calculated from the design value of the shape of the edge E to be measured. Alternatively, as shown in FIG. 5, a user may input at least three or more points in the vicinity of the edge E to be measured, and a line smoothly connecting these points may be defined as a curve L. At this time, the curve L can be calculated by performing fitting using a polynomial, spline calculation, or the like. The spline calculation is an arithmetic method for obtaining a curve L by smoothly connecting a plurality of input points, and is attributed to obtaining a spline function. A spline function is defined by different polynomial curves of order n at most within a small section, and they are connected as smoothly as possible to each other.
[0026]
Then, when measuring the coordinates of the edge E, a plurality of scanning start points on the defined curve L are set, the optical probe 2 is scanned in a direction substantially orthogonal to the curve L from each point, and a threshold value is set. The coordinates of the edge E are obtained by comparing the above. At this time, if the curve L intersects the edge E to be measured in the range of the edge E to be measured, scanning with the optical probe 2 becomes difficult, and the coordinates of the edge E cannot be measured accurately. In such a case, it is necessary to increase the number of input points used to calculate the curve L or increase the number of polynomial terms to be fitted. Thereby, it is possible to easily measure the coordinates of the edge E made of an arbitrary curve with a small number of input values.
[0027]
Further, the edges E1 and E2 in the test surface S1 and the surface shape measurement values in the effective area B of the test surface S1 are measured with the same measurement coordinate system using the same non-contact three-dimensional measuring machine, and the surface shape is measured. By fitting the data to the design formula and performing coordinate transformation on the measurement data of the edges E1 and E2, the design coordinate system (local coordinate system) (X, Y) of each surface with respect to the outer shape as shown in FIG. ) Can be accurately obtained. Thereby, since the amount of eccentricity in the surface of the test surface S1 can be known, it can be reflected in elucidation of the cause of optical performance deterioration, correction, and the like.
[0028]
In FIG. 6, only the left and right edges E1 and E2 in the drawing are measured, but all edge portions or only necessary edge portions may be measured. The shapes of the edges E1 and E2 are not limited to curves, and may be arcs or straight lines. At this time, by combining with the conventional edge measurement method (FIG. 3), it becomes possible to measure efficiently. In FIG. 6, the curve L1 is a curve defined along the edge E1, and the curve L2 is a curve defined along the edge E2.
[0029]
Here, examples of the test object O in the present invention include an arbitrary optical element such as a prism or a mold used when manufacturing such an optical element by plastic molding, glass molding or the like. Such optical elements include not only spherical surfaces and rotationally symmetric aspheric surfaces, but also those using a refractive surface or a reflective surface for a cylindrical surface, an anamorphic surface, or a free-form surface. Here, there are various defining formulas as a method of expressing a free-form surface, and one example is defined by the following formula. The Z axis of this defining formula is the axis of the free-form surface.
[0030]

Here, the first term of the equation (a) is a spherical term, and the second term is a free-form surface term.
[0031]
In the spherical term,
c: vertex curvature
k: Conic constant (conical constant)
r = √ (X ² + Y ² )
It is.
[0032]
The free-form surface term is

However, C _j (J is an integer of 2 or more) is a coefficient.
[0033]
In general, the free-form surface does not have a symmetric surface in both the XZ plane and the YZ plane, but by setting all odd-order terms of X to 0, the plane of symmetry is parallel to the YZ plane. Is a free-form surface with only one. Further, by setting all odd-numbered terms of Y to 0, a free-form surface having only one symmetry plane parallel to the XZ plane is obtained.
[0034]
Next, a method for calculating the relative eccentricity of the plurality of surfaces constituting the edge by measuring the edges formed by the boundaries of the plurality of surfaces for each surface constituting the edges and combining them will be described. Here, for example, a design coordinate system (local coordinate system) for each surface of an optical element such as a free-form surface, a prismatic surface, a cylindrical surface, an anamorphic surface, an aspherical surface, and a spherical surface that are arranged eccentrically with respect to each other. The case where the relative positional relationship of (coordinate system) is calculated | required is shown.
[0035]
With reference to FIG. 7, a method of calculating the amount of eccentricity of the test surfaces S1 and S2 sharing one edge E of the test object (prism) O will be described. FIG. 7A is a perspective view showing a state in which the shape of the surface and the shape of the edge are measured by scanning the optical probe 2 with respect to the test surface S1 of the test object O, and FIG. FIG. 7C is a top view of the test surface S1, and FIG. 7C is a perspective view showing how the shape of the surface and the shape of the edge are measured by scanning the optical probe 2 with respect to the test surface S2. d) is a top view of the test surface S2.
[0036]
First, as shown in FIG. 6, the shape measurement value of the edge E and the surface shape measurement value in the effective areas B <b> 1 and B <b> 2 are the same using the same noncontact three-dimensional measuring machine for each of the test surfaces S <b> 1 and S <b> 2. Measurement is performed in the measurement coordinate system, fitting is performed by applying the surface shape data to the design values of the test surfaces S1 and S2, and the coordinate conversion is also performed on the measurement data of the edge E, so that the test surfaces S1 and S2 are obtained. The position of the edge E with respect to the local coordinate systems A1 and A2 to be expressed is obtained.
[0037]
Further, coordinate conversion is performed so that the data of the edge E measured for each surface coincide. At this time, the edge measurement value of one surface may be fixed, and the edge measurement value of the other surface may be decentered and shifted so that the surface edges overlap. For the analysis, a least square method may be used so that the measured values of both edges are on substantially the same curve.
[0038]
By executing the coordinate transformation obtained by this analysis also on the local coordinate systems A1 and A2 representing the test surfaces S1 and S2, the local coordinate systems on the plurality of surfaces S1 and S2 as shown in FIG. It becomes possible to clarify the relative positional relationship between A1 and A2. 8 shows an example in which only two test surfaces are included in the edge E. However, the present invention is not limited to this, and the relative position of the local coordinate system having three or more test surfaces is the same. It can be obtained by a technique. By these methods, since the relative eccentricity in a plurality of surfaces can be known, it is possible to reflect in the elucidation and correction of the cause of optical performance deterioration.
[0039]
Next, as shown in FIG. 9, a reference element (reference block) D is used in which the amount of eccentricity of the two test surfaces S1 and S2 sharing the common edge E is fixed to the element of the test object O. The calculation method will be described. FIG. 9A shows the shape of the test surface S1 and the edge e of the reference surface H by scanning the optical probe 2 with respect to the test surface S1 of the test object O and one reference surface H of the reference block D. FIG. 9B is a top view of the test surface S1, and FIG. 9C is an optical probe with respect to the test surface S2 and another reference surface H of the reference block D. FIG. 9 is a perspective view showing a state in which the shape of the test surface S2 and the shape of the edge e of the reference surface H are measured by scanning 2, and FIG. 9D is a top view of the test surface S2.
[0040]
The test object O is integrally fixed and held by the reference block D and the holder C, and the shapes of the reference surfaces H of the reference block D and the corresponding test surfaces S1 and S2 are the same non-contact 3 Enables measurement in the same measurement coordinate system using a dimension measuring machine.
[0041]
The surface shape data of the test surfaces S1 and S2 are fitted to the design values of the test surfaces S1 and S2, respectively, so that the local coordinate system A1 representing the test surfaces S1 and S2 is represented. Find the position of A2.
[0042]
In addition, when calculating the local coordinate system, the coordinates that minimize the residual when the design shape of each test surface is coordinate-transformed and the surface shape data and each design shape that has undergone coordinate conversion are superimposed Analysis of optimization to obtain transformation parameters The You may go.
[0043]
The shape of the reference block D is configured so that at least two edges e or one edge e and the position of a point whose coordinates are known can be measured simultaneously with the measurement of the test surfaces S1 and S2 of the test object O. Is desirable. At this time, at least two edges e used for measurement or one edge e and a point whose coordinates are known must be such that the position of each reference plane H of the reference block D can be defined from the measurement results. There is. And at least Of test object O Each measurement Face and The angle and length of the edge e to be measured at the same time are measured in advance with high accuracy and are known, and the relative position of each reference plane H defined by the measurement value is known from these known data. Keep it like that.
[0044]
The shape of the reference block D may be, for example, a polyhedral shape having a configuration in which the test surfaces S1 and S2 are approximated by a plane in accordance with the shape of the test object O. At this time, it is good also as a polyhedron comprised by the plane defined by local coordinate system A1, A2 of the design value of each to-be-tested surface S1, S2. When the reference block D is configured by a plane, the measurement can be simplified with high accuracy, and the analysis can be easily performed on other curved surfaces. Further, by adopting such a shape, each reference surface H of the reference block D and each of the test surfaces S1 and S2 corresponding to the reference surface D can be held substantially horizontally, and using the same measurement coordinate system, Measurement can be easily performed.
[0045]
Next, a measurement method will be described. First, as shown in FIG. 9, the shape measurement value of the edge e of the reference block D and the surface shape measurement value in the effective areas B1 and B2 of the test object O are the same for each of the test surfaces S1 and S2. Measurement is performed in the same measurement coordinate system using a non-contact three-dimensional measuring machine. Further, as will be described later, when a mark v whose coordinate position is known in the local coordinate system on the surface to be tested is processed in an area other than the effective area, similarly, the same measurement coordinate is used. Measure with system.
[0046]
For example, when it is necessary to face the test surfaces S1 and S2 upward due to the measurement machine, as shown in FIGS. 9A to 9D, the test object O is rotated for each measurement of each surface. Need to be held. Therefore, the test object O and the reference block D are rotated and fixed in a desired direction by using a rotation mechanism (not shown). At this time, the relative positional relationship between the test object O and the reference block D is not changed by rotation.
[0047]
The reference block D performs measurement of items such that the reference plane position can be obtained from the measurement value of the reference block D in correspondence with the known measurement value. FIG. 9 shows an example in which two edges e are measured for each reference plane H. From these data, the relative position of the reference plane H of the reference block D can be obtained.
[0048]
For example, in the measurement of the edge of each reference surface H of the test surfaces S1 and S2, the edge to be measured is assumed to include a ridge line that becomes the boundary of the reference surface constituting the reference block D, and further different edge portions are measured. The position of the ridgeline and the position of the plane representing the reference plane may be obtained for each reference plane. Here, the position of the ridge line means the result of deriving the position of at least one vertex of the side of the reference plane indicated by the ridge line and a linear expression representing the ridge line. In order to calculate the vertex, an intersection point may be obtained from the measurement result of the shape of two edges forming the vertex in the reference plane. Also, for example, a point whose positional relationship with the vertex of the reference plane is known can be measured, and the position of the ridgeline and the reference plane are similarly expressed from the measurement result of the point and the measurement result of the edge that becomes the ridgeline. The position of the plane may be obtained. At this time, a least square method or the like may be used for the analysis so that the measured value of the edge is substantially linear. Further, the position of each reference plane can be obtained by associating the derived plane with the known measurement values (side length or side-side angle) of the reference block D. However, if the position of the plane and the position of the ridge line are known as the respective reference surface positions, the relative values of the test surfaces S1 and S2 described later can be derived.
[0049]
Furthermore, by measuring the shapes of the test surfaces S1 and S2 in the same measurement coordinate system, the relative positions of the reference surface H and the test surfaces S1 and S2 can be grasped for each test surface S1 and S2. It becomes possible.
[0050]
As will be described later, when a mark v whose coordinate position is known in the local coordinate system on the surface to be tested is processed in an area other than the effective area, By measuring the mark position, the relative position of the mark v on the reference surface H and the test surfaces S1 and S2 can be grasped for each test surface S1 and S2.
[0051]
Further, the position of each surface obtained from the measurement value of each reference surface H is converted in correspondence with the known data of the reference surface H constituting the reference block D, and the relative positional relationship of each surface is matched. At this time, it is necessary to perform coordinate transformation such as shift and rotation.
[0052]
For example, coordinate conversion is performed so that the positions of the ridge lines derived from the measurement results of the respective reference planes H including the same ridge line are spatially matched, and thereby the ridge line is included with the spatially matched ridge line as a rotation axis. A parameter for performing coordinate conversion of rotation may be calculated so that each plane corresponds to a known value of the reference element (an angle formed by each reference plane forming the same ridge line with the reference element). As a result, the relative position of each reference plane H can be obtained using the known angle of the reference block D.
[0053]
By performing these conversions also on the shape measurement data of each test surface S1, S2 measured in the same measurement coordinate system as each reference surface, the relative position of each test surface S1, S2 is calculated. Is possible.
[0054]
Further, as will be described later, when a mark v whose coordinate position is known when processed in a local coordinate system on the test surface is processed in an area other than the effective area, similarly to the surface shape, By measuring each mark position in the same measurement coordinate system as each test surface and performing the same conversion on these mark positions, the relative position of each mark v on the test surfaces S1 and S2 is determined. It is possible to calculate.
[0055]
By performing the coordinate transformation obtained by this analysis also on the local coordinate systems A1 and A2 of the test surfaces S1 and S2, the local coordinate systems A1 and S2 on the plurality of surfaces S1 and S2, as shown in FIG. It becomes possible to clarify the relative positional relationship of A2. Although FIG. 10 shows an example of only two surfaces constituting the edge E, the present invention is not limited to this, and the relative positions of three or more local coordinate systems can be obtained by a similar method. . In this embodiment, one reference surface of the reference block D is measured simultaneously for one test surface. However, one reference surface for two or more test objects may be used depending on the shape of the test object. May be measured simultaneously. By these methods, since the relative eccentricity in a plurality of surfaces can be known, it is possible to reflect in the elucidation and correction of the cause of optical performance deterioration.
[0056]
In addition, it is possible to manufacture an optical element with high accuracy by performing manufacturing based on the result of this method.
[0057]
Referring to FIGS. 11 to 14, the eccentric amount of each optical element of the test object O in which a plurality of optical elements (for example, prisms) are arranged on the frame W is fixed to the frame W that holds the test object O. A method of calculating using a reference element (reference block) D will be described. At this time, the test object O can measure at least one surface shape constituting each optical element in a certain direction. In FIGS. 11 to 14, the prism P <b> 1 and the prism P <b> 2 are disposed within the frame W, and the surface S <b> 1-1 constituting the prism P <b> 1 and the S <b> 2-2 constituting the prism P <b> 2 are exposed to the outside. By doing so, each surface can be measured.
[0058]
FIG. 11 shows the shape of the test surface S1-1 and the edge e of the reference surface H by scanning the optical probe 2 with respect to the test surface S1-1 of the prism P1 and one reference surface H of the reference block D. FIG. 12 is a top view of the test surface S1-1, and FIG. 13 is a plan view of the test surface S2-1 of the prism P2 and another reference surface H of the reference block D. FIG. 14 is a top view of the test surface S2-1. FIG. 14 is a top view of the test surface S2-1, and shows the state of measuring the shape of the test surface S2-1 and the shape of the edge e of the reference surface H by scanning the optical probe 2. is there.
[0059]
The test object O is integrally fixed and held by the reference block D and the holder C, and the shapes of the reference surfaces H of the reference block D and the test surfaces S1-1 and S2-1 corresponding thereto are respectively shown. Measurement can be performed in the same measurement coordinate system using a non-contact three-dimensional measuring machine. The surface shape data of the test surfaces S1-1 and S2-1 are fitted to the design values of the test surfaces S1-1 and S2-1, respectively, so that the test surfaces S1-1 and S2 are applied. The positions of the local coordinate systems A1-1 and A2-1 representing -1 are obtained in advance.
[0060]
In addition, when calculating the local coordinate system, the coordinates that minimize the residual when the design shape of each test surface is coordinate-transformed and the surface shape data and each design shape that has undergone coordinate conversion are superimposed Optimization analysis for obtaining conversion parameters may be performed.
[0061]
The shape of the reference surface H is such that at least two edges e or one edge e and coordinates are known simultaneously with the measurement of the test surfaces S1-1 and S2-1 of the prisms P1 and P2 of the test object O. It is desirable to configure so that the position of can be measured. At this time, at least two edges e used for measurement or one edge e and a point whose coordinates are known must be such that the position of each reference plane H of the reference block D can be defined from the measurement results. There is. Further, at least the angle and length of the edge e to be measured simultaneously with each measurement surface H are measured in advance with high accuracy, and the relative position of each reference surface H defined by measurement values from these known data. Make sure you understand.
[0062]
The shape of the reference block D may be, for example, a polyhedron shape having a surface that approximates each of the test surfaces S1-1 and S2-1 in accordance with the shape of the prisms P1 and P2 of the test object. Good. At this time, the shape of a polyhedron having a plane defined by the local coordinate systems A1-1 and A2-1 of the design values of the test surfaces S1-1 and S2-1 may be used. When the reference block D is configured by a plane, the measurement can be simplified with high accuracy, and the analysis can be easily performed on other curved surfaces. The material may be glass or metal. Manufacturing with these materials enables processing with high surface accuracy, so that the angle of the reference surface can be measured with high accuracy. In addition, since a sharp edge shape can be formed, the reference surface can be measured with high accuracy.
[0063]
Next, a measurement method will be described. First, as shown in FIGS. 11 to 14, the shape of the edge e of the reference block D and the test surfaces of the prisms P <b> 1 and P <b> 2 that are the test object O for each test surface S <b> 1-1 and S <b> 2-1. The surface shapes in the effective areas B1-1 and B2-1 of S1-1 and S2-1 are measured in the same measurement coordinate system using the same non-contact three-dimensional measuring machine. At this time, for example, the test surfaces S1-1 and S2-1 are set so that the test surfaces S1-1 and S2-1 are substantially perpendicular to the incident direction of the optical probe 2 due to the measuring machine. When it is necessary to face upward, as shown in FIGS. 11 to 14, it is necessary to rotate and hold the test object O for each measurement of each surface. Therefore, the test object O and the reference block D are rotated and fixed in a desired direction by using a rotation mechanism (not shown). At this time, the relative positional relationship between the test object O and the reference block D is not changed by rotation.
[0064]
At this time, if the reference block D is in the shape of a polyhedron having a plane defined by the local coordinate systems A1-1 and A2-1 of the design values of the test surfaces S1-1 and S2-1, the test is performed. When measuring the test surfaces S1-1 and S2-1 of the object O and the reference surface of the reference block D, the test surfaces S1-1 and S2-1 of the test object O and the reference surface of the reference block D However, since it can be arranged so as to be substantially perpendicular to the incident direction of the optical probe 2 at the same time, the measurement can be easily performed.
[0065]
The reference block D performs measurement of items that allow the reference surface position to be obtained from the measurement value of the reference block D by corresponding to a known measurement value such as the length of the side of each reference surface. FIGS. 11 to 14 show an example in which two edges are measured for each reference plane H. FIG. From these measurement data, the position of the reference surface H of the reference block D can be obtained.
[0066]
For deriving the position of the reference plane H, a method similar to the example shown in FIG. 9 can be used. Thereby, the position of each reference plane H of the reference block D can be obtained. Further, by measuring the shapes of the test surfaces S1-1 and S2-1 in the same measurement coordinate system, the relative positions of the reference surface H and the test surfaces S1-1 and S2-1 are determined. -1 and S2-1 can be grasped.
[0067]
Further, the position of each surface obtained from the measurement value of each reference surface H is converted in correspondence with the known data of the reference surface H constituting the reference block D, and the relative positional relationship of each surface is matched. At this time, it is necessary to perform coordinate transformation such as shift and rotation.
[0068]
In addition, in order to obtain | require the relative position of each surface, the method similar to the example shown in FIG. 9 can be used. By performing these conversions on the test surfaces S1-1 and S2-1 measured in the same measurement coordinate system as the reference surfaces, the relative positions of the test surfaces S1-1 and S2-1 are determined. It becomes possible to grasp.
[0069]
By performing the coordinate transformation obtained by this analysis also on each of the local coordinate systems A1-1 and A2-1 of the test surfaces S1-1 and S2-1, different optical elements as shown in FIG. It is possible to obtain the relative positional relationship between the local coordinate systems A1-1 and A2-1 on the plurality of surfaces S1-1 and S2-1 constituting the (prisms P1 and P2).
[0070]
11 to 14 show examples including two prisms P1 and P2. However, the present invention is not limited to this, and the relative positions of three or more optical elements are obtained by the same method. It becomes possible. In addition, the optical element (prisms P1 and P2) is shown as an example in which the test surface is one surface. However, the present invention is not limited to this, and a plurality of surfaces may be measured for each element. Thus, it becomes possible to obtain more eccentricity and dimensional information. In this case, the number of reference surfaces of the reference block may be increased as necessary. Moreover, although the example which calculates | requires the relative position of a different prism as a to-be-tested object was shown, the to-be-tested object is not limited to this, A lens, the metal mold | die of an optical element, a stop, and the frame W may be used. For example, in the case of a frame, the surface shape of the surface to be measured may be measured, but it is only necessary to measure what can determine the position of the frame from the measured value. H may be measured in the same measurement coordinate system.
[0071]
Further, if the relative position of the local coordinates of each surface of each prism is obtained by the method shown in FIG. 9 before being incorporated in the frame W, for example, the local coordinate system of each surface constituting the prism for each prism is obtained. Since the relative positions are known, the test surfaces S1-1 and S2-1 of the prisms P1 and P2 shown in FIG. 15 are made to correspond to the positions of the local coordinate systems A1-1 and A2-1. Thus, as shown in FIG. 16, surfaces other than S1-1 (S1-2, S1-3) constituting the prism P1 and surfaces other than S2-1 (P2-2, S2-3) constituting the prism P2. The relative positions of the local coordinate systems (A1-2, A1-3, A2-2, A2-3) of each surface can be obtained. By these methods, the relative amount of eccentricity in a plurality of elements can be known, and therefore it can be reflected in the elucidation and correction of the cause of optical performance deterioration. In addition, it is possible to manufacture an optical system with high accuracy by performing manufacturing based on the result of this method.
[0072]
Next, with reference to FIG. 17, for example, the eccentric amounts of S3-1 and S3-2 of the test surface arranged on the front and back surfaces of the test object O such as an aspheric lens are measured. A method of calculating using a reference element (reference block) D fixed to the above will be described. FIG. 17A shows the shape of the test surface S3-1 and the reference by scanning the optical probe 2 with respect to the test surface S3-1 of the aspherical lens O and one reference surface H of the reference block D. FIG. 17B is a perspective view showing the state of measuring the shape of the edge e of the surface H, FIG. 17B is a perspective view of the test surface S3-1, and FIG. 17C is the test surface S3-2 and the reference block. It is a perspective view which shows a mode that the shape of S3-2 of a to-be-tested surface and the shape of the edge e of the reference surface H are measured by scanning the optical probe 2 with respect to the other reference surface H of D. FIG.
[0073]
The test object O is integrally fixed and held by the reference block H and the holder C, and the shapes of the reference surfaces H of the reference block D and the corresponding test surfaces S3-1 and S3-2 are respectively set. Measurement can be performed in the same measurement coordinate system using a non-contact three-dimensional measuring machine. The surface shape data of the test surfaces S3-1 and S3-2 are fitted to the design values of the test surfaces S3-1 and S3-2, respectively, so that the test surfaces S3-1 and S3 are applied. The positions of the local coordinate systems A1-1 and A2-1 representing -2.
[0074]
The shape of the reference block H is the position of a point whose coordinates are known at least two edges e or one edge e simultaneously with the measurement of the test surfaces S3-1 and S3-2 of the aspherical lens of the test object O. It is desirable to be able to measure. At this time, at least two edges e used for measurement or one edge e and a point whose coordinates are known are such that the position of H can be defined on each reference plane of the reference block D from the measurement results. There is a need. Further, at least the angle and length of the edge e to be measured simultaneously with each measurement surface H are measured in advance with high accuracy, and the relative position of each reference surface H defined by measurement values from these known data. Make sure you understand.
[0075]
The shape of the reference block D may be, for example, a rectangular parallelepiped having a thickness of about each of the test surfaces S3-1 and S3-2 of the test object O. At this time, two edges may be measured at each reference plane with the opposing plane as a reference plane. At this time, the length of each side of the reference block D, the parallelism between the reference surfaces, the angle with the surface adjacent to the reference surface (orthogonality), the height between the reference surfaces, etc. are measured with high accuracy. It is desirable.
[0076]
Next, a measurement method will be described. First, as shown in FIG. 17, the shape of the edge e of the reference block D for each of the test surfaces S3-1 and S3-2 and the effectiveness of the test surfaces S3-1 and S3-2 of the test object O are shown. The surface shape of the area is measured in the same measurement coordinate system using a non-contact three-dimensional measuring machine. At this time, for example, when it is necessary to face the test surfaces S3-1 and S3-2 upward so that the test surface is substantially perpendicular to the direction in which the optical probe 2 is incident due to the measurement machine, As shown in FIGS. 17A to 17C, it is necessary to rotate and hold the test object O by approximately 180 ° for each measurement of each surface. At this time, the relative positional relationship between the test object O and the reference block D is not changed by rotation.
[0077]
The reference block D performs measurement of items that allow the reference surface position to be obtained from the measurement value of the reference block D by corresponding to a known measurement value such as the length of the side of each reference surface. FIG. 17 shows an example in which two edges are measured for each reference plane H. From these measurement data, the position of the reference surface H of the reference block D can be obtained. For deriving the position of the reference plane H, a method similar to the example shown in FIG. 9 can be used. Thereby, the position of each reference plane H of the reference block D can be obtained. Further, by measuring the shapes of the test surfaces S3-1 and S3-2 in the same measurement coordinate system, the relative positions of the reference surface H and the test surfaces S3-1 and S3-2 are determined for each test surface S3. -1, S3-2 can be grasped.
[0078]
Further, the position of each surface obtained from the measurement value of each reference surface H is converted in correspondence with the known data of the reference surface H constituting the reference block D, and the relative positional relationship of each surface is matched. When the reference block D is a substantially rectangular parallelepiped, for example, the coordinates of the vertices of each reference plane and the inclination of each reference plane calculated by measurement may be associated with the relative position and parallelism data of the vertices of known data. In FIG. 17, the shape of the edge of the ridge line that becomes the boundary of the reference surface is not measured, but analysis is possible if the relative positional relationship between the edges of each reference surface is known. By performing these conversions on the test surfaces S3-1 and S3-2 measured in the same measurement coordinate system as the reference surfaces, the relative positions of the test surfaces S3-1 and S3-2 can be determined. It becomes possible to grasp. Further, in addition to the opposing reference planes, the relative positions of the reference planes that are in contact with both reference planes are measured. An analysis for deriving the relative position may be performed.
[0079]
By performing the coordinate transformation obtained by this analysis also on each of the local coordinate systems A3-1 and A3-2 of the test surfaces S3-1 and S3-2, as shown in FIG. It becomes possible to clarify the relative positional relationship between the local coordinate systems A3-1 and A3-2 on the surfaces S3-1 and S3-2 constituting the aspheric lens. In addition, since the relative position of each test surface can be found by analysis, it is possible to obtain information such as lens thickness in addition to decentration. Furthermore, it is possible to manufacture an optical element with high accuracy by performing manufacturing based on such information.
[0080]
If the shape of the test surface is aspherical, the position of the local coordinates can be obtained by fitting to the design value from the surface shape data of the test surface, but if the shape of the test surface is spherical, Even if this method is used, the position of the local coordinate system cannot be determined. In such a case, a mark M may be provided on the surface to be measured so that the position of the local coordinate system can be obtained by simultaneously measuring the shape of the surface to be measured. In this case, the relative position of the mark M with respect to the local coordinate system is known, so that the position of the local coordinate system can be derived from the measurement results of a plurality of marks. The shape of the mark is not limited as long as the above analysis is possible. For example, the shape of the mark may be a cross as shown in FIG. 18A or a cylinder as shown in FIG. If the above analysis is possible from the outer shape, it is not necessary to process the mark or the like. For example, in the case of an oval outer shape as shown in FIG. 18 (c), if the relative position of the local coordinate system can be derived from the shape of the linear portion of the outer shape, the position of the vertex with respect to the arc portion, etc. An analysis for deriving a local coordinate system of the surface to be measured may be performed from the measurement result.
[0081]
In addition, when the shape of the test surface is a spherical surface, for example, the spherical center position can be calculated by fitting the surface shape data with a sphere, so the relative spherical center position of each surface can also be obtained. is there. At that time, in the measurement of the spherical test surface, it is desirable to measure, for example, a mark and an outer shape as shown in FIG. 18 in the same coordinate system.
[0082]
The measurement of the outer shape is not limited to the oval shape as shown in FIG. 18C, and may be a normal circular outer shape. In that case, the position of the center of the outer shape can be obtained by fitting the measurement result of the outer shape with a circle. Thereby, for example, it is possible to obtain the relative position of the sphere center of each surface with respect to a reference position such as the mark position or the outer shape center. Similarly, when an aspherical surface is included, the approximate spherical center position can be obtained in addition to the local coordinate system by fitting the aspherical surface shape data with a spherical surface. By using this method in this way, it becomes possible to measure the eccentricity of the specimen having a rotationally symmetric optical surface.
[0083]
Next, referring to FIG. 20, the amount of eccentricity of each optical element of the test object O in which a plurality of optical elements (for example, lenses, prisms) are arranged in a frame W as shown in FIG. A method of calculating using a reference element (reference block) D fixed to the frame W will be described. At this time, the test object O is an optical system in which the optical axis is bent by a prism. In addition, the test object O enables measurement of the surface shape of a plurality of surfaces in a certain direction. Therefore, it is desirable that the frame W has a configuration in which the frame W and a member constituting the measuring machine, for example, a lens that irradiates the optical probe 2 (not shown) do not collide when the surface to be measured is measured.
[0084]
In FIG. 19, the first lens group G1, the prism P3, and the second lens group G2 are arranged in the frame W, and the surface SL1-1 and the prism P3 that form the lens L1-1 of the first lens group G1 are arranged. The surface SP3-2 to be formed and the surface SL2-5 to form the lens L2-5 of the second lens group G2 are on the outside, and each surface can be measured by irradiating the optical probe 2. Yes.
[0085]
FIG. 20A shows the shape of the test surface SL1-1 and the edge e of the reference surface H by scanning the optical probe 2 with respect to the test surface SL1-1 and one reference surface H of the reference block D. FIG. 20B is a perspective view showing the state of measuring the shape of the test surface by scanning the optical probe 2 with respect to the test surface SP3-2 and the other reference surface H of the reference block D. FIG. 20C is a perspective view showing a state of measuring the shape of SP3-2 and the shape of the edge e of the reference surface H. FIG. 20C shows the light to the test surface SL2-5 and the other reference surface H of the reference block D. It is a perspective view which shows a mode that the shape of to-be-tested surface SL2-5 and the shape of the edge e of the reference plane H are measured by scanning the probe 2. FIG.
[0086]
The test object O is integrally fixed and held by the reference block H and the holder C, and each reference surface H of the reference block D and each test surface SL1-1, SP3-2, SL2-5 corresponding thereto. The shape of each can be measured in the same measurement coordinate system using a non-contact three-dimensional measuring machine. The surface shape data of each of the test surfaces SL1-1, SP3-2, and SL2-5 are fitted to the design values of the test surfaces SL1-1, SP3-2, and SL2-5, respectively. The positions of the local coordinate systems AL1-1-1, AP3-2, and AL2-5-2 representing the test surfaces SL1-1, SP3-2, and SL2-5 are obtained in advance.
[0087]
In addition, when calculating the local coordinate system, the coordinates that minimize the residual when the design shape of each test surface is coordinate-transformed and the surface shape data and each design shape that has undergone coordinate conversion are superimposed An optimization analysis for obtaining a conversion parameter may be performed.
[0088]
The shape of the reference block D is such that at least two edges e or 1 are measured simultaneously with the measurement of the test surfaces SL1-1, SP3-2, SL2-5 of the lenses L1-1, L2-5 and the prism P3 of the test object O. It is desirable to be able to measure the position of one edge e and the point whose coordinates are known. At this time, at least two edges e used for measurement or one edge e and a point whose coordinates are known are such that the position of H can be defined on each reference plane of the reference block D from the measurement results. There is a need. Further, at least the angle and length of the edge e to be measured simultaneously with each measurement surface H are measured in advance with high accuracy, and the relative position of each reference surface H defined by measurement values from these known data. Make sure you understand.
[0089]
The shape of the reference block D may be, for example, a polyhedron shape having a surface that approximates each of the test surfaces SL1-1, SP3-2, and SL2-5 in accordance with the shape of the test object O. Good. At this time, the shape of the polyhedron having a plane defined by the local coordinate systems AL1-1-1, AP3-2, AL2-5-2 of the design values of the test surfaces SL1-1, SP3-2, SL2-5 It may be.
[0090]
Next, a measurement method will be described. First, as shown in FIG. 20, the shape of the edge e of the reference block D and the test surfaces SL1-1 and SP3 of the test object O for each test surface SL1-1, SP3-2 and SL2-5. -2, and the surface shape in the effective area of SL2-5 are measured in the same measurement coordinate system using a non-contact three-dimensional measuring machine. At this time, the test surfaces SL1-1, SP3 are set so that the test surfaces SL1-1, SP3-2, SL2-5 are substantially perpendicular to the direction in which the optical probe 2 is incident, for example, due to the measurement machine. -2, SL2-5 needs to face upward, as shown in FIGS. 20A to 20C, it is necessary to rotate and hold the test object O for each measurement of each surface. Therefore, the test object O and the reference block D are rotated and fixed in a desired direction by using a rotation mechanism (not shown). At this time, the relative positional relationship between the test object O and the reference block D is not changed by rotation.
[0091]
The reference block D performs measurement of items such that the reference plane position can be obtained from the measurement value of the reference block D by corresponding to a known measurement value such as the length of the side of each reference plane. FIG. 20 shows an example in which two edges are measured for each reference plane H. From these measurement data, the position of the reference surface H of the reference block D can be obtained. For deriving the position of the reference plane H, a method similar to the example shown in FIG. 9 can be used. Thereby, the position of each reference plane H of the reference block D can be obtained. Further, by measuring the shapes of the test surfaces SL1-1, SP3-2, and SL2-5 in the same measurement coordinate system, the relative relationship between the reference surface H and the test surfaces SL1-1, SP3-2, and SL2-5 is measured. It is possible to grasp the correct position for each test surface SL1-1, SP3-2, SL2-5.
[0092]
Further, the position of each surface obtained from the measurement value of each reference surface H is converted in correspondence with the known data of the reference surface H constituting the reference block D, and the relative positional relationship of each surface is matched. At this time, it is necessary to perform coordinate transformation such as shift and rotation. In addition, in order to obtain | require the relative position of each surface, the method similar to the example shown in FIG. 9 can be used. By performing these conversions on the test surfaces SL1-1, SP3-2, SL2-5 measured in the same measurement coordinate system as the reference surfaces, the test surfaces SL1-1, SP3-2, SL2 It becomes possible to grasp the relative position of −5.
[0093]
The coordinate transformation obtained by this analysis is also performed on each of the local coordinate systems AL1-1-1, AP3-2, and AL2-5-2 of the test surfaces SL1-1, SP3-2, and SL2-5. Thus, as shown in FIG. 22, the relative positional relationship of the local coordinate systems AL1-1-1, AP3-2, and AL2-5-2 on the plurality of planes SL1-1, SP3-2, and SL2-5 is obtained. It becomes possible.
[0094]
Further, before being incorporated in the frame W, it is desirable to obtain the relative position of the local coordinates of each surface of each optical element arranged in the test object for each group or prism, for example. FIGS. 21A to 21C show a lens group G1 (consisting of lenses L1-1 and L1-2), a prism P3, and a lens group G2 (lenses L2-1, L2-2, and L2) that constitute the test object. -3, L2-4, and L2-5), the results of measuring the eccentricity of each surface are shown. The decentering in the lens groups G1 and G2 can be derived using an existing measuring machine, and the decentering of the prism P3 can also be derived by the method shown in FIG. 9, for example. From the derived result, the relative position of the local coordinate system for each surface can be known. As shown in FIGS. 21A to 21C, for example, in the lens group G1, the lens L1-1 is configured. The relative positions of the surface local coordinate systems AL1-1-1 and AL1-1-2 and the surface local coordinate systems AL1-2-1 and AL1-2-2 constituting the lens L1-2 are known. For example, in the prism P3, the relative positions of the local coordinate systems AP3-1, AP3-2, and AP3-3 on the surface constituting the prism P3 are known. Further, in the lens group G2, local coordinate systems AL2-1-1 and AL2-1-2 of the surfaces constituting the lens L2-1, and local coordinate systems AL2-2-1 of the surfaces constituting the lens L2-2, AL2-2-2, the local coordinate system AL2-3-2 of the surface constituting the lens L2-3, the local coordinate systems AL2-4-1 and AL2-4-2 of the surface constituting the lens L2-4 The relative positions of the local coordinate systems AL2-5-1 and AL2-5-2 of the surfaces constituting the lens L2-5 are known.
[0095]
At this time, since the relative position of each surface in the local coordinate system is known for each lens group and prism, as shown in FIG. 20, the test surfaces SL1-1 of the lenses L1-1, L2-5, and the prism P3 are used. Surfaces other than the test surface SL1-1 constituting the lens group G1 by combining relative to the local coordinate systems AL1-1-1, AP3-2, and AL2-5-2 of SP3-2 and SL2-5 And local coordinate systems (AL1-1-1, AL1-1-2, AL1-2-1, each surface including a surface other than SP3-2 constituting the prism P3 and a surface other than SL2-5 constituting the lens group G2. AL1-2-2, AP3-1, AP3-2, AP3-3, AL2-1-1, AL2-1-2, AL2-2-1, AL2-2-2, AL2-3-2, AL2- 4-1, AL2-4-2, AL2-5-1, AL2-5 It can be used to determine the relative position of the two). By these methods, the relative amount of eccentricity in a plurality of elements can be known, and therefore it can be reflected in the elucidation and correction of the cause of optical performance deterioration. In addition, it is possible to manufacture an optical system with high accuracy by performing manufacturing based on the result of this method.
[0096]
It should be noted that when measuring the shape of each test surface, it is shown that a rotation mechanism (not shown) is used when it is necessary to arrange the test surface so that the test surface faces substantially upward. Since it is only necessary to fix the specimen without changing the relative position, the rotation mechanism may not be used as long as such a holding is possible. For example, the posture to be arranged may be changed and fixed.
[0097]
Incidentally, in order to measure the three-dimensional surface shape of the test surface 1 (S1, S2) of the test object O by scanning the optical probe 2 of the non-contact three-dimensional measuring machine as shown in FIG. 1 or FIG. Since it is difficult to continuously scan the entire surface of the inspection surface 1 without any problems due to measurement time or the like, the measurement is usually performed at sampled points in the measurement surface. Regarding the movement of the optical probe 2 between the jumping points, conventionally, as shown in FIG. 24B, the height of the optical probe 2 in the state (1) in which the height is measured at a specific position is kept as it is. Move horizontally while keeping, and measure the next position in the state of (2). However, in such movement of the optical probe 2 between the measurement points, when the gradient of the test surface 1 is steep, the objective lenses 15 and 24 that emit the optical probe 2 come into contact with the test surface 1 to be measured. There are cases where the surface 1 or the objective lenses 15 and 24 are damaged, or the distance from the surface 1 to be inspected becomes too far and the reflected light cannot be taken into the objective lenses 15 and 24 again to be measured.
[0098]
Therefore, in the measurement method of the present invention, as shown in FIG. 24A, the optical probe 2 in the state (1) whose height is measured at a specific position is moved upward by a predetermined minute distance α. Set to the state of (2), then move away from the design shape of the test surface 1 by a small distance α along the shape and bring it to the next measurement position to the state of (3). The height of the next measurement position is measured. Here, the movement from the state {circle around (2)} to the state {circle around (3)} may be performed at the shortest distance or along the plane of the design value + α. By adopting such a movement method of the optical probe 2, it is possible to avoid both the objective lenses 15 and 24 coming into contact with the surface 1 to be measured and the fact that the reflected light cannot be captured. Note that the minute distance α is 0 or more, and can be arbitrarily set within a range in which the reflected light from the test surface 1 can be captured by the objective lens of the non-contact three-dimensional measuring machine.
[0099]
Next, a measurement method using a sample in which the mark v has been processed at the coordinate position designated in the local coordinate system on the surface to be inspected O will be described.
[0100]
FIG. 25 is an example of a sample in which the mark v has been processed. As shown in FIG. 25, in the measurement method of the present invention, a sample in which the mark v is processed at a coordinate position designated by the local coordinate system on the test surface of the test object O may be used. In this case, the mark position can be measured in the same measurement coordinate system when measuring the surface shape of the test surface, and the position of the local coordinate system of the test surface can be obtained from the measured value of each mark position. Accordingly, when derivation of the local coordinate system is performed, analysis such as fitting and optimization is not necessary, and thus analysis is facilitated.
[0101]
The mark shape is not limited in shape and number as long as the above analysis is possible. For example, the shape may be a cylinder, a prism, a cone, a pyramid, a cross, an L shape, an intersection, a point, or the like. . Moreover, as long as measurement can be performed with high accuracy, either a convex shape or a concave shape may be used. Further, it is desirable that the mark processing position is processed into an area other than the effective areas B1 and B2 of the optical system. For example, when the mark shape is a cylinder, the mark position may be the center position of the outer circle of the upper surface. In the case of a cross shape, the position of the intersection of each straight line may be used.
[0102]
Each mark need not be in the same plane as the test surface as long as the mark shape can be measured in the same measurement coordinate system when measuring the surface shape of the test surface. It may be on the side.
[0103]
FIG. 26 is another example of a sample in which the mark v has been processed. As shown in FIG. 26, in the measuring method of the invention, a sample in which the mark v is processed at the coordinate position designated by the local coordinate system on the surface to be inspected O may be used. In this case, the mark position can be measured in the same measurement coordinate system when measuring the surface shape of the test surface, and the position of the local coordinate system of the test surface can be obtained from the measured value of each mark position. Therefore, the positions of the local coordinate systems A1 and A2 indicating the test surfaces are obtained from the measurement results of the marks v on the test surfaces S1 and S2.
[0104]
Further, as shown in FIG. 27, the shape of the reference block may be a prism whose bottom surface is a polyhedron. In this case, when the test object O is held, the test surfaces S1 and S2 and at least one surface constituting the reference block D are held so as to be substantially horizontal. Here, each of the test surfaces S1 and S2 and the surface of the reference block D that is substantially horizontal are referred to as a reference surface H.
[0105]
In FIG. 27, since the surfaces H1 and H2 constituting the reference block are held substantially horizontally with the test surfaces S1 and S2, respectively, they are used as the reference surfaces H of the test surfaces. With such a configuration, it is not necessary to manufacture a reference block according to the shape of the test object O, and the same reference block can be used for general purposes. In FIG. 27, the shape of the reference block is an octagonal column, but the shape is not limited to this as long as the above-described holding is possible.
[0106]
Next, referring to FIG. 28 to FIG. 31, the eccentric amount of each optical element of the test object O in which a plurality of optical elements (for example, prisms) are arranged on the frame W is determined as the frame W holding the test object O. A second method of calculating using a reference element (reference block) D fixed to will be described. At this time, the test object O can measure a mark position processed in at least one surface constituting each optical element in a certain direction.
[0107]
The mark shape is not limited in shape and number as long as the above measurement is possible. For example, the shape includes a cylinder, a prism, a cone, a pyramid, a cross, an L shape, an intersection, a point, and the like. . Moreover, as long as measurement can be performed with high accuracy, either a convex shape or a concave shape may be used. Further, it is desirable that the mark processing position is processed into an area other than the effective areas B1-1 and B2-1 of the optical system.
[0108]
In FIGS. 28 to 31, the prism P1 and the prism P2 are arranged in the frame W, and the marks v processed on the surface S1-1 constituting the prism P1 and the surface S2-1 constituting the prism P2, respectively, are on the outside. Each mark shape can be measured by irradiating the optical probe, and as a result, the position of the mark v can be obtained.
[0109]
FIG. 28 shows the mark shape and the reference surface H on the test surface S1-1 by scanning the optical probe with respect to the test surface S1-1 of the prism P1 and one reference surface H of the reference block D. FIG. 29 is a top view of the test surface S1-1, and FIG. 30 shows the test surface S2-1 of the prism P2 and another reference surface H of the reference block D. FIG. 31 is a perspective view showing a state in which the shape of the mark on the test surface S2-1 and the shape of the edge e of the reference surface H are measured by scanning the optical probe, and FIG. 31 shows the test surface S2-1. It is a top view.
[0110]
The test object O is fixed and held integrally by the reference block H and the holder C, and the marks on the reference surfaces H of the reference block D and the corresponding test surfaces S1-1 and S2-1. Each of the shapes can be measured with the same measurement coordinate system using a non-contact three-dimensional measuring machine, and the mark position is obtained from the measured value of the mark shape.
[0111]
The shape of the reference block is the position of a point whose coordinates are known at least two edges e or one edge e simultaneously with the measurement of the test surfaces S1-1 and S2-1 of the prisms P1 and P2 of the test object O. It is desirable to be able to measure. At this time, it is necessary to be able to define the position of H on each reference plane of the reference block D based on at least two edges e used for measurement or one edge e and a point whose coordinates are known. Furthermore, at least the angle and length of the edge e measured simultaneously with each measurement surface H are measured in advance with high accuracy, and the relative position of each reference surface H defined by measurement values from these known data is determined. Make sure you understand.
[0112]
Next, the measurement method will be described. 11 to 14, the shape of the edge e of the reference block D and the prisms P <b> 1 and P <b> 2 that are the test object O are measured for each test surface S <b> 1-1 and S <b> 2-1. Each mark v on the test surfaces S1-1 and S2-1 is measured in the same measurement coordinate system using the same non-contact three-dimensional measuring machine. At this time, for example, when it is necessary to face the test surfaces S1-1 and S2-1 upward so that the test surfaces are substantially perpendicular to the direction in which the optical probe is incident, for example, in relation to a measuring machine. As shown in FIGS. 28 to 31, it is necessary to rotate and hold the test object O for each measurement of each surface. Therefore, the test object O and the reference block D are rotated and fixed in a desired direction by using a rotation mechanism (not shown). At this time, the relative positional relationship between the test object O and the reference block D is not changed by rotation.
[0113]
It should be noted that the method similar to the example shown in FIG. 9 can be used for deriving the position of the reference plane H, whereby the position of each reference plane H of the reference block D can be obtained. Further, by measuring the position of each mark v on the test surfaces S1-1 and S2-1 in the same measurement coordinate system, the test surface H and the marks on the test surfaces S1-1 and S2-1 are measured. It becomes possible to grasp the relative position of v for each of the test surfaces S1-1 and S2-1.
[0114]
For this mark v, for example, by using the same method shown in FIG. 26, that is, by measuring the mark position using the same measurement coordinate system together with the shape measurement of each surface, the same analysis is performed. Together with the surface shape of each surface, the relative position of the mark can be calculated. Therefore, it is possible to derive the relative position between the surface shape of the test surface and the mark position.
[0115]
Further, the position of each surface obtained from the measurement value of each reference surface H is converted in correspondence with the known data of the reference surface H constituting the reference block D, and the relative positional relationship of each surface is matched. At this time, it is necessary to perform coordinate transformation such as shift and rotation.
[0116]
The coordinate transformation obtained by this analysis is also applied to each mark v on each of the test surfaces S1-1 and S2-1, so that the marks on the test surfaces S1-1 and S2-1 can be obtained. It becomes possible to grasp the relative position.
[0117]
Further, if the mark position of each test surface of each prism and the relative position of the local coordinate system are obtained by, for example, the method shown in FIG. 26 before being incorporated in the frame W, each prism constituting each prism is configured. The relative position between the surface mark and the local coordinate system is known.
[0118]
In addition, if the mark is processed into a coordinate position specified in the local coordinate system on the surface to be measured, the position of the local coordinate system can be easily calculated from the measurement result of each mark position. Therefore, for example, by the method shown in FIG. 26, the relative positions of the plurality of marks v on each surface constituting the prism and the local coordinate system can be determined for each prism.
[0119]
However, even when the coordinate position of each mark in the local coordinate system of the test surface is unknown, the analysis can be performed. In this case, when measuring the position of the mark, the shape of the test surface is measured using the same measurement coordinate system. For the surface shape data, the position of the local coordinate system representing the test surface is obtained by performing an optimization analysis that minimizes the residual when superimposed on each design value shape, for example, by coordinate transformation. Therefore, the relative position of the local coordinate system and the mark v is known for each surface. Therefore, for example, by the method shown in FIG. 26, the relative position of each mark v is known together with the surface shape of each surface constituting the prism and the local coordinate system for each prism. In this case, even if the mark v is not on the same surface as the test surface, the same analysis is possible if the mark position can be measured in the same measurement coordinate system.
[0120]
Therefore, by making these results correspond to the positions of the same marks on the test surfaces S1-1 and S2-1 of the prisms P1 and P2 shown in FIG. 30, as shown in FIG. The local coordinate system (A1-) of the surfaces other than S1-1 (S1-2, S1-3) constituting P1 and the surfaces other than S2-1 (S2-2, S2-3) constituting the prism P2. 2, A1-3, A2-2, A2-3) can also be obtained.
[0121]
In the above analysis, for example, for each corresponding test surface, coordinate conversion is performed between the mark position measured in FIG. You may perform the analysis of optimization which calculates a coordinate transformation parameter so that the sum total of distance may become the minimum. Furthermore, the coordinate transformation derived by this analysis may be performed on the local coordinate system of the test surface or the local coordinate system of another test surface whose relative positional relationship with the mark position is known.
[0122]
By these methods, the relative amount of eccentricity in a plurality of elements can be known, and therefore it can be reflected in the elucidation and correction of the cause of optical performance deterioration. In addition, it is possible to manufacture a highly accurate optical system by performing manufacturing based on the result of this method.
[0123]
In this method, it is necessary to measure the mark on the test surface before and after being incorporated into the frame W. However, since it does not require analysis such as fitting or optimization when calculating the mark position, it is incorporated into the frame W. It is possible to reduce the influence of measurement errors before and after.
[0124]
Further, in the above analysis, a case has been shown in which a single test surface with a mark is used for each prism. However, the present invention is not limited to this. For example, one prism may be formed on a plurality of test surfaces. Analysis may be performed using a certain mark. In this case, since the correspondence information of the mark position is increased, the accuracy of the relative eccentricity in the calculated plurality of elements is improved.
[0125]
The above three-dimensional coordinate measuring method of the present invention can be configured as follows, for example.
[0126]
[1] A curve approximately along the edge is set in the vicinity of the edge on the test object, and a coordinate value at a position determined based on the edge is detected by scanning in a direction substantially orthogonal to the curve. A three-dimensional coordinate measuring method.
[0127]
[2] The three-dimensional coordinate measuring method as described in 1 above, wherein the curve is defined by a polynomial.
[0128]
[3] The three-dimensional coordinate measuring method according to the above item 1, wherein the curve is a curve defined by selecting a plurality of points in the vicinity of an edge and gently connecting the plurality of points.
[0129]
[4] The three-dimensional coordinate measurement method according to the above item 3, wherein spline calculation is used when the plurality of points are smoothly connected.
[0130]
[5] The three-dimensional coordinate measuring method according to the above item 1, wherein at least one of the surfaces constituting the test object is a free-form surface.
[0131]
[6] The three-dimensional coordinate measuring method as described in 5 above, wherein the test object is an optical element in which at least one of the constituent surfaces is a free-form surface.
[0132]
[7] The three-dimensional coordinate measuring method according to the above item 5, wherein the test object is a mold for manufacturing an optical element in which at least one of the constituent surfaces is a free-form surface.
[0133]
[8] With respect to a test object composed of a plurality of surfaces, a three-dimensional surface shape and an edge shape are measured for each surface, and the measured values of the plurality of surfaces are made to correspond to the measured values of the common edge. A three-dimensional coordinate measurement method characterized by obtaining a relative positional relationship.
[0134]
[9] The relative positional relationship between the plurality of surfaces is obtained by associating the measured values of the common edges measured for different surfaces so as to be on substantially the same curve. 9. The three-dimensional coordinate measuring method according to 8.
[0135]
[10] A local coordinate system for each surface is obtained by applying the surface shape measurement value measured for each surface and the surface design value, and the common edge measurement value measured on one surface and the other surface 3. The relative positional relationship of the local coordinate system in a plurality of planes is obtained by performing coordinate transformation that associates the measured values of the common edge with each other on the local coordinate system. Dimensional coordinate measurement method.
[0136]
[10-1] By optimizing the coordinate transformation parameter that minimizes the residual when the surface shape measurement value measured for each surface and the coordinate-transformed design shape of each test surface are superimposed Then, a local coordinate system for each surface is obtained, and coordinate conversion is performed on the local coordinate system so that the measured value of the common edge measured on one surface corresponds to the measured value of the common edge measured on the other surface. 10. The three-dimensional coordinate measurement method according to 9 above, wherein the relative positional relationship of the local coordinate system on a plurality of surfaces is obtained.
[0137]
[11] The three-dimensional coordinate measurement method as described in 9 above, wherein at least one of the surfaces constituting the test object is a free-form surface.
[0138]
[12] The three-dimensional coordinate measuring method according to [11], wherein the test object is an optical element in which at least one of the constituent surfaces is a free-form surface.
[0139]
[13] The three-dimensional coordinate measuring method as described in 11 above, wherein the test object is a mold for manufacturing at least one of the constituent surfaces is a free-form surface.
[0140]
[14] A test object composed of a plurality of surfaces and a reference element whose relative positional relationship is known are relatively fixed, and the three-dimensional surface shape of each test surface of the test object Measure the shape of the reference element in the same measurement coordinate system, convert the coordinates of each reference surface of the reference element according to the surface shape data obtained from the measured value of the edge shape of the reference element, and perform the same coordinate conversion for each A three-dimensional coordinate measurement method characterized in that a relative positional relationship among a plurality of test surfaces of a test object is obtained by performing measurement on a three-dimensional surface shape measurement value of the test surface.
[0141]
[14-1] A test object composed of a plurality of surfaces, a reference element whose relative positional relationship between the reference surfaces is known, and the test object are relatively fixed, and each test surface of the test object The three-dimensional surface shape and the reference element are measured in the same measurement coordinate system, the reference surface coordinates of each reference element are converted corresponding to the surface shape data obtained from the measurement values of the reference element, and the same coordinate conversion is performed. A three-dimensional coordinate measuring method characterized in that a relative positional relationship between a plurality of test surfaces of a test object is obtained by performing also on the position of a mark on each test surface.
[0142]
[14-2] In a test object composed of a plurality of surfaces, a reference element having a known relative positional relationship between the reference surface and the test object are relatively fixed, and each test surface of the test object Measure the position of the mark above and the reference element in the same measurement coordinate system, convert the reference surface coordinates of each reference element according to the surface shape data obtained from the measurement value of the reference element, and the same coordinate conversion The three-dimensional coordinate measurement method is characterized in that the relative positional relationship between the plurality of test surfaces of the test object is obtained by performing the above on the position of the mark on each test surface.
[0143]
[14-3] In a test object composed of a plurality of surfaces, a reference element having a known relative positional relationship between the reference surface and the test object are relatively fixed, and each test surface of the test object Measure the position of the mark above and the reference element in the same measurement coordinate system, convert the reference surface coordinates of each reference element according to the surface shape data obtained from the measured value of the edge shape of the reference element, A three-dimensional coordinate measuring method, wherein the relative coordinate relationship between a plurality of test surfaces of a test object is obtained by performing the same coordinate conversion on the positions of the marks on each test surface.
[0144]
[15] In a test object configured by using a plurality of elements composed of a plurality of surfaces, a reference element having a known relative positional relationship between the reference surface and the test object are relatively fixed, and the test object The three-dimensional shape of at least one surface of each element and the reference element are measured in the same coordinate system, and the coordinates of each reference surface of the reference element are set in correspondence with the surface shape data obtained from the measured value of the edge shape of the reference element. A three-dimensional coordinate measurement characterized by obtaining a relative positional relationship of at least one surface constituting each element by performing the same coordinate conversion on the three-dimensional surface shape measurement value of each test surface. Method.
[0145]
[15-1] In a test object configured by using a plurality of elements each having a plurality of surfaces, a reference element whose relative positional relationship is known with a reference surface is relatively fixed, and the test object is fixed. The three-dimensional shape of at least one surface of each element of the inspection object and the reference element are measured in the same measurement coordinate system, and the coordinates of the reference surface of each reference element are determined in accordance with the surface shape data obtained from the measurement values of the reference element. A three-dimensional coordinate measurement characterized by obtaining a relative positional relationship of at least one surface constituting each element by performing the same coordinate transformation on the three-dimensional shape measurement value of each test surface. Method.
[0146]
[15-2] In a test object configured by using a plurality of elements composed of a plurality of surfaces, a reference element whose relative positional relationship is known with a reference surface is relatively fixed, and the test object is fixed. Each reference element is measured in accordance with the surface shape data obtained from the measurement value of the reference element by measuring the position of the mark on the at least one test surface of each element of the inspection object and the reference element in the same measurement coordinate system. By converting the coordinates of the reference plane of the reference plane and performing the same coordinate transformation on the position of the mark on each test surface to obtain the relative positional relationship between at least one plane constituting each element. A three-dimensional coordinate measuring method.
[0147]
[15-3] In a test object configured by using a plurality of elements each composed of a plurality of surfaces, a reference element whose relative positional relationship is known with a reference surface is relatively fixed, and the test object is fixed. Measure the position of a mark on at least one test surface of each element of the inspection object and the reference element in the same measurement coordinate system, and correspond to the surface shape data obtained from the measured value of the edge shape of the reference element. By converting the coordinates of the reference surface of each reference element and performing the same coordinate conversion on the mark position of each test surface, the relative positional relationship of at least one surface constituting each element is obtained. A three-dimensional coordinate measuring method characterized by the above.
[0148]
[16] The above-mentioned 14, 14-3, 15 characterized in that the shape of the reference element is obtained by measuring at least two edges of the reference element or one edge and the position of a point whose coordinates are known. 15-3, The three-dimensional coordinate measuring method in any one of.
[0149]
[17] The three-dimensional coordinate measurement method as described in 16 above, wherein the angle formed by each reference plane constituting the reference element is known.
[0150]
[18] The three-dimensional coordinate measuring method as described in 17 above, wherein the edge to be measured of the reference plane includes a ridge line formed by each reference plane.
[0151]
[19] Deriving a position of a ridge line composed of each reference surface and a position of a plane representing the reference surface from a measurement result of at least two edges of the reference surface or a point whose coordinates are known with one edge. 19. The three-dimensional coordinate measuring method according to 18 above.
[0152]
[20] The coordinates are converted so that the positions of the ridge lines derived from the measurement results of the respective reference planes including the same ridge line are spatially coincident, and the ridge line is thus obtained with the spatially coincident ridge line as a rotation axis. 20. The three-dimensional coordinate measurement method as described in 19 above, wherein a parameter for performing coordinate conversion is calculated so that each plane to be included corresponds to a known value of the reference element.
[0153]
[21] The three-dimensional coordinate measurement method as described in 20 above, wherein the transformation of the coordinates is performed on a measurement result of each test surface measured in the same measurement coordinate system as each reference surface.
[0154]
[22] The three-dimensional coordinates according to [20], wherein the transformation of the coordinates is performed on a local coordinate system obtained from a measurement result of each test surface measured in the same measurement coordinate system as each reference surface. Measuring method.
[0155]
[23] A local coordinate system for each surface is obtained by applying the surface shape measurement value measured for each surface and the design value of the surface, and the three-dimensional surface of each test surface is obtained with respect to the obtained local coordinate system. 14, 14-3, 15, 15, wherein the relative positional relationship of the local coordinate system on a plurality of test surfaces of the test object is obtained by performing coordinate conversion performed on the shape measurement value. The three-dimensional coordinate measuring method in any one of -3.
[0156]
[23-1] Each cover Using the sample that has processed the mark at the coordinate position specified in the local coordinate system on the inspection surface, the local coordinate system is obtained from the mark position measured for each surface, and for the obtained local coordinate system, The above 14 or 14-2 or 15 or 15, wherein the relative positional relationship of the local coordinate system on a plurality of test surfaces of the test object is obtained by performing coordinate conversion performed on each reference surface. -2 described in the three-dimensional coordinate measuring method.
[0157]
[23-2] By optimizing the coordinate transformation parameter that minimizes the residual when the surface shape measurement value measured for each surface and the coordinate-transformed design shape of each test surface are superimposed By calculating the local coordinate system for each surface and performing coordinate transformation performed on each reference surface with respect to the obtained local coordinate system, the relative coordinate system of the plurality of test surfaces of the test object The three-dimensional coordinate measuring method according to the above 14 or 14-2 or 15 or 15-2, characterized in that a simple positional relationship is obtained.
[0158]
[24] The three-dimensional coordinate measurement method according to any one of the above 14, 14-3, 15, and 15-3, wherein at least one of the surfaces constituting the test object is a free-form surface.
[0159]
[25] The three-dimensional coordinates according to any one of the above 14, 14-3, 15, and 15-3, wherein the test object is an optical element in which at least one of the constituent surfaces is a free-form surface. Measuring method.
[0160]
[26] The three-dimensional coordinate measuring method as described in 24 above, wherein the test object is a mold for manufacturing at least one of the constituent surfaces is a free-form surface.
[0161]
[27] The three-dimensional coordinate measuring method according to any one of the above 14, 14-3, 15, and 15-3, wherein a reference plane constituting the reference element is configured by a plane.
[0162]
[27-1] The three-dimensional coordinate measuring method according to the above 14 or 14-3 or 15 or 15-2, wherein the reference element is a prism having a polygonal bottom surface.
[0163]
[28] The three-dimensional coordinate measurement method according to any one of 14, 14-3, 15, and 15-3, wherein the shape of the reference element is a polyhedron having a configuration in which a surface to be measured is approximated by a plane. .
[0164]
[29] The shape of the reference element is a polyhedron composed of a plane defined by the local coordinate system of the surface to be tested, according to any one of the above 14, 14-3, 15, 15-3 3D coordinate measurement method.
[0165]
[30] The three-dimensional coordinate measuring method according to any one of the above 14, 14-3, 15, and 15-3, wherein the reference element is made of glass.
[0166]
[31] The three-dimensional coordinate measurement method according to any one of the above 14, 14-3, 15, and 15-3, wherein the reference element is made of metal.
[0167]
[32] The three-dimensional coordinate measurement method according to any one of the above 14, 14-3, 15, and 15-3, wherein the test object and the reference element are held in a jig having a rotation mechanism.
[0168]
[33] Measurement is performed using a non-contact optical probe, and the non-contact optical probe at a specific measurement point is moved to a predetermined measurement point by moving the non-contact optical probe from a specific measurement point to the next measurement point. Move it away from the test surface by a minute distance, then move away from the design value of the test surface by that minute distance along the design value, and move to the next measurement point. 33. The three-dimensional coordinate measurement method according to any one of items 1 to 32, wherein measurement points are measured.
[0169]
[34] The three-dimensional coordinate measurement method as described in 15 or 15-3 above, wherein the relative positional relationship between the surfaces constituting the elements constituting the test object is known.
[0170]
[35] The three-dimensional coordinate measuring method as described in 16 above, wherein a test object configured by using a plurality of elements composed of a plurality of surfaces is held by a frame.
[0171]
[36] The three-dimensional coordinate measurement method as described in 16 above, wherein the frame for holding the test object is configured to irradiate the test surface with light.
[0172]
[37] The three-dimensional coordinate measuring method according to 15 or 15-3, wherein the test object has a rotationally symmetric optical surface.
[0173]
[38] The three-dimensional coordinate measurement method according to item 15 or 15-3, wherein the test object has a bent optical axis.
[0174]
[39] In a three-dimensional coordinate measurement method for measuring a three-dimensional coordinate of a surface to be measured with a non-contact optical probe, when the non-contact optical probe is moved from a specific measurement point to the next measurement point, at a specific measurement point The non-contact optical probe is moved away from the test surface by a predetermined minute distance, and then moved away from the design value of the test surface by the minute distance along the design value to the next measurement point. A three-dimensional coordinate measurement method, characterized in that the next measurement point is measured in a state of being moved.
[0175]
[40] A measuring instrument using the three-dimensional coordinate measuring method described in any one of 1 to 39 above.
[0176]
[41] Measured by the three-dimensional coordinate measurement method described in any one of 1 to 39 above.
[0177]
【The invention's effect】
As is apparent from the above description, according to the three-dimensional coordinate measurement method of the present invention, the relative position of the design coordinate system (local coordinate system) of each test surface with respect to the outer shape can be accurately obtained, It becomes possible to clarify the relative positional relationship of the local coordinate system on the test surface. Therefore, the relative amount of eccentricity on a plurality of test surfaces can be known, and can be reflected in elucidation of the cause of optical performance degradation, correction, and the like.
[Brief description of the drawings]
FIG. 1 is a diagram schematically showing a configuration of an autofocus type non-contact three-dimensional measuring machine that can be used in the present invention.
FIG. 2 is a diagram schematically showing the configuration of a confocal microscope type non-contact three-dimensional measuring machine that can be used in the present invention.
FIG. 3 is a diagram for explaining a method of measuring a circular edge portion.
FIG. 4 is a diagram for explaining a method of measuring the shape of an edge according to the present invention.
FIG. 5 is a diagram for explaining a method of setting a curve indicating a scanning start point when measuring an edge shape.
FIG. 6 is a diagram for explaining a method of obtaining a relative position of a local coordinate system of each surface with respect to an outer shape based on the present invention.
FIG. 7 is a diagram for explaining a method of calculating the amount of eccentricity of a plurality of test surfaces that share one edge of a test object based on the present invention.
8 is a diagram showing a local coordinate system of a plurality of surfaces whose relative positional relationship has been clarified by the method of FIG. 7;
FIG. 9 is a diagram for explaining a method of calculating the eccentric amounts of a plurality of test surfaces using a reference block fixed to the test object according to the present invention.
10 is a diagram showing a local coordinate system of a plurality of surfaces whose relative positional relationship has been clarified by the method of FIG. 9;
FIG. 11 is a diagram illustrating a method for deriving the amount of eccentricity of a test surface that constitutes each element of a plurality of elements using a reference block fixed to the test object according to the present invention. It is a perspective view which shows a mode that the shape of the to-be-tested surface and the shape of the edge of a reference surface are measured by scanning an optical probe with respect to the to-be-tested surface of an element and one reference surface of a reference block.
12 is a top view of the test surface of FIG. 11. FIG.
13 shows how the shape of the test surface and the shape of the edge of the reference surface are measured by scanning the optical probe with respect to the test surface of another element of FIG. 11 and another reference surface of the reference block. It is a perspective view shown.
14 is a top view of the test surface of FIG.
15 is a diagram showing a relative position in a local coordinate system of a surface constituting each element of a plurality of elements whose relative positional relationship has been clarified by the method of FIGS. 11 to 14; FIG.
FIG. 16 clearly shows the relative positional relationship by using, as a test object, an optical system composed of elements in which the decentering of the surfaces constituting each element is derived in advance by the method of FIGS. It is a figure which shows the relative position of the local coordinate system of each surface which comprises the several element which became.
FIG. 17 is a diagram for explaining a method of deriving the amount of eccentricity of a test surface that constitutes an aspheric lens using a reference block fixed to a test object according to the present invention.
FIG. 18 is a diagram showing the outer shape of a mark or test object processed to derive a local coordinate system of the test surface when the test surface is a spherical surface by the method of FIG. 17;
FIG. 19 is a diagram for explaining a test object including two lens groups and a prism.
20 shows a method for deriving the amount of eccentricity of the test surface constituting the test object comprising the two lens groups and the prism shown in FIG. 19 using a reference block fixed to the test object according to the present invention. It is a figure for demonstrating.
FIG. 21 is a diagram for explaining a local coordinate system of each surface constituting each element (lens, prism) for the two lens groups and prisms shown in FIG. 20;
22 is a diagram showing a relative position in a local coordinate system of a surface constituting each element of a plurality of groups and prisms whose relative positional relationship has been clarified by the method of FIG. 20;
FIG. 23 shows a plurality of relative positional relationships clarified by using, as a test object, an optical system composed of elements in which the decentering of the surfaces constituting each element is derived in advance by the method of FIG. It is a figure which shows the relative position of the local coordinate system of each surface which comprises an element.
FIG. 24 is a diagram for explaining a method of moving an optical probe in the present invention.
FIG. 25 is a diagram for explaining a method of obtaining a relative position of a local coordinate system of each surface with respect to a mark position by using an example of a sample in which a mark v has been processed.
FIG. 26 is a diagram for explaining a method of calculating the amount of eccentricity of a plurality of test surfaces using a reference block fixed to a test object, using another example of a sample in which the mark v has been processed. It is.
FIG. 27 is a diagram for explaining a method of calculating the eccentricity of a plurality of test surfaces using a reference block of a prism whose bottom surface is fixed to a test object according to the present invention.
FIG. 28 is a diagram for explaining a method for deriving the amount of eccentricity of a test surface constituting each element of a plurality of elements using a reference block fixed to the test object according to the present invention. It is a perspective view which shows a mode that the position of the mark on the to-be-tested surface and the shape of a reference surface are measured by scanning an optical probe with respect to the to-be-tested surface of an element and one reference surface of a reference block.
29 is a top view of the test surface of FIG. 28. FIG.
30 measures the position of the mark on the test surface and the shape of the reference surface by scanning the optical probe with respect to the test surface of another element in FIG. 28 and another test surface of the reference block. It is a perspective view which shows a mode to do.
31 is a top view of the test surface of FIG. 30. FIG.
[Explanation of symbols]
O ... Subject
S1, S2, S1-1, S2-1, S1-2, S1-3, S2-2, S2-3, S3-1, S3-2, SL1-1, SP3-2, SL2-5 ... test surface
P1, P2, P3 ... Prism
L1-1, L1-2, L2-1, L2-2, L2-3, L2-4, L2-5 ... Lens
G1, G2 ... Lens group
A1, A2, A1-1, A2-1, A1-2, A1-3, A2-2, A2-3, A3-1, A3-2, AL1-1-1, AL1-1-2, AL1- 2-1, AL1-2-2, AP3-1, AP3-2, AP3-3, AL2-1-1, AL2-1-2, AL2-2-1, AL2-2-2, AL2-3- 2, AL2-4-1, AL2-4-2, AL2-5-1, AL2-5-2 ... Local coordinate system
W ... Frame
M ... Reference mark
E, E1, E2, e ... Edge
L, L1, L2 ... curve
B, B1, B2, B1-1, B2-1 ... Effective area
D ... Reference block
H, H1, H2 ... Reference plane
C ... Retaining tool
F ... yen
v ... mark
1 ... surface to be measured
2 ... Optical probe
11 ... Laser
12, 13, 14 ... mirror
15 ... Objective lens
16 ... Movement mechanism
17 ... Optical position detection device
18 ... XY stage
21 ... Light source
22 ... 1st pinhole
23 ... Half mirror
24 ... Objective lens
25 ... 2nd pinhole
26. Photodetector
27 ... Moving mechanism
28 ... XY stage

Claims (9)

And the test object consists of a plurality of faces, the relative positional relationship of the reference plane is fixed and known reference element, said reference corresponding to the 3-dimensional face shape and each side of each face of the object The edge of the reference plane of the element is measured in the same measurement coordinate system, and the coordinates of each reference plane obtained from the measured value of the edge are converted so as to correspond to the relative positional relationship of each known reference plane. By aligning the relative positional relationships of the positions of the reference planes obtained from the measured values of the edge, and performing the same coordinate conversion on the three-dimensional surface shape measurement values of each surface of the test object, A three-dimensional coordinate measuring method, characterized in that a relative positional relationship between each surface of the two is obtained. 2. The three-dimensional coordinate measuring method according to claim 1, wherein each reference plane of the reference element is a polyhedron configured by approximating a test surface with a plane. The position of a ridge line composed of each reference plane and the position of a plane representing the reference plane are derived from the measurement result of at least two edges of the reference element or a point whose coordinates are known with one edge. Item 3. A three-dimensional coordinate measurement method according to Item 2. By optimizing the coordinate transformation parameters that minimize the residual when superimposing the three-dimensional surface shape measurement value measured for each surface of the surface to be tested and the design shape of each surface to be coordinate transformed , Obtaining a local coordinate system for each surface, and performing a coordinate transformation performed on the position of the reference surface also on the local coordinate system, thereby obtaining a relative positional relationship of the local coordinate systems on a plurality of surfaces. The three-dimensional coordinate measuring method according to any one of claims 1 to 3. For a test object composed of a plurality of surfaces, a three-dimensional surface shape and an edge shape are measured for each surface, and a common edge shape measurement value is associated with the local coordinates on the plurality of surfaces. Is a three-dimensional coordinate measurement method for obtaining the relative positional relationship between each surface, and a local coordinate system for each surface is obtained by applying the three-dimensional surface shape measurement value measured for each surface and the design value of the surface. common by the shape measurement of a common edge, measured by the shape measurement and the other surface of the edge is the correspondence to ride on substantially the same curve, the local seat of the plurality of surfaces measured at One surface three-dimensional measuring how to and obtains the relative positional relationship of the target. A curve that substantially follows the edge is set in the vicinity of the edge on the test object, and a coordinate value at a position determined based on the edge is detected by scanning in a direction substantially orthogonal to the curve. Item 6. The three-dimensional coordinate measurement method according to Item 5 . For a test object composed of a plurality of surfaces, a three-dimensional surface shape and an edge shape are measured for each surface, and a common edge shape measurement value is associated with the local coordinates on the plurality of surfaces. Is a three-dimensional coordinate measurement method for obtaining a relative positional relationship between the three-dimensional surface shape measurement value measured for each surface and the coordinate-transformed design shape of each surface to be measured, and a residual. by but optimized for obtaining the coordinate conversion parameter that minimizes obtains the local coordinate system of each surface, common edge shape measured at a common shape measurement edges and other surfaces measured in one plane measurements by to the correspondence to ride on substantially the same curve, the three-dimensional measuring how to and obtains the relative positional relationship of the local coordinates in the plurality of surfaces. By measuring a three-dimensional surface shape and an edge shape for each surface with respect to a test object composed of a plurality of surfaces and associating a common edge shape measurement value with each other, Is a three-dimensional coordinate measuring method for obtaining an accurate positional relationship, wherein a curve substantially along the edge is set in the vicinity of the edge on the test object, scanned in a direction substantially orthogonal to the curve, and determined based on the edge A three-dimensional coordinate measuring method, comprising: detecting a coordinate value of a selected position. The relative positional relationship between the plurality of surfaces is obtained by associating the shape measurement values of the common edge measured for each different surface so as to be on substantially the same curve. 9. The three-dimensional coordinate measuring method according to 8.

Images