JP2004049638A - Endoscope apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an endoscope apparatus having a reduced diameter of an insertion part and high measurement accuracy which alleviates fatigue when stereoscopic viewing is carried out. <P>SOLUTION: This endoscope apparatus 2 comprises a pair of objective-optical systems; an imaging device 28 on which images are formed through an image transmitting optical system for transmitting optical images from the objective-optical systems; and a means for correcting multi-distortions; at a tip 21 of the endoscope 2. The means for correcting the multi-distortions corrects the multi-distortions formed due to overlapping of a first distortion caused by the pair of the objective-optional systems and a second distortion caused by an image transmitting optical device 27. This endoscope apparatus 2 stores table data to correct the multi-distortions and optical data such as a focal distance in a memory card 33 and carries out excellent measurement by using the optical data. <P>COPYRIGHT: (C)2004,JPO

Description

により求められる。
【００７７】
続いて、ステップＳ２６において、補正画像範囲を設定し、本ルーチンを終了する。なお、このサブルーチンの画像補正処理は、左右２画像に対して行い、原画像を計測画像に変換する。
【００７８】
次に、図８のステップＳ１４にてコールされるサブルーチンの計測処理について、図１０のフローチャートにより説明する。
【００７９】
まず、ステップＳ１７１において、ＬＣＤ５に表示されるメニュー上にて３つの計測方式を選択する。その１つは、深さ計測であり、点と平面間の距離を計測する場合、例えば、図１１の斜視図に示すようにパイプ４１等の内面の腐食の深さｈ（図１２部分拡大図参照）を計る場合に用いられる。
【００８０】
この計測方式を選択するとステップＳ１７２にジャンプする。他の１つは、長さ計測であり、２点間の距離を計測する場合、例えば、図１３の斜視図に示すように立体的被計測物であるパイプ４２の２点の３次元座標を求め、２点間の距離ｄを求める場合に適用される。
【００８１】
この計測方式を選択するとステップＳ１９０にジャンプする。他の１つは、くぼみ計測であり、点と直線間の距離を計測する場合、例えば、図１４の斜視図に示すように、タービンブレード４３の欠損等の大きさ（欠け部分の寸法ｈ）等を計測する場合に適用される。この計測方式が選択されるとステップＳ２００にジャンプする。
【００８２】
上記ステップＳ１７２にジャンプした場合、ステップＳ１７３〜ステップＳ１７５において、基準平面設定用の３点の基準点ＰＴを設定する。例えば、図１１に示す被計測物のパイプ５１の場合、測定基準点ＰＴ１，ＰＴ２，ＰＴ３を設定する。そして、後述するサブルーチン３次元座標解析処理（図１５（Ａ）参照）により上記３点の３次元座標を求める。
【００８３】
ステップＳ１７６で点ＰＴ指示値をインクリメントし、ステップＳ１７７で上記３点が作る平面の深さの基準平面として設定する。図１１の場合、点ＰＴ１，ＰＴ２，ＰＴ３を含む平面Ｍ１を設定する。さらに、ステップＳ１７８で測定点、例えば、ＰＴ４を設定し　、Ｓ１７９でサブルーチン３次元座標解析処理（図１５（Ａ）参照）を呼び出して実行して測定点の座標を求める。図１２の場合、点ＰＴ４の座標を求める。
【００８４】
ステップＳ１８０において、基準平面から測定点間の距離、図１２の場合、測定点ＰＴ４から前記基準平面Ｍ１までの腐食の深さを与える距離ｈを演算し、本ルーチンを終了する。
【００８５】
次に、ステップＳ１９０にジャンプした場合は、ステップＳ１９１で第１測定点を設定し、ステップＳ１９２でサブルーチン３次元座標解析処理を実行し、上記第１測定点の座標を求める。さらに、ステップＳ１９３で第２測定点を設定し、ステップＳ１９４でサブルーチン３次元座標解析処理を実行し、上記第２測定点の座標を求める。その後、ステップＳ１９５において上記第１，第２測定点の座標から上記２点間の距離ｄを演算し、本サブルーチンを終了する。
【００８６】
なお、上記の処理において、被計測物が図１３に示すパイプ４２であって、第１，第２測定点がＰＴ５，ＰＴ６であった場合、上記計測により上記２測定点間の距離ｄが求められる。
【００８７】
さらに、ステップＳ２００にジャンプした場合は、ステップＳ２０１でくぼみの一端に第１基準点を設定し、ステップＳ２０２でサブルーチン３次元座標解析処理を実行し、上記第１基準点の座標を求める。さらに、ステップＳ２０３でくぼみの他端に第２基準点を設定し、ステップＳ２０４でサブルーチン３次元座標解析処理を実行し、上記第２基準点の座標を求める。ステップＳ２０５において上記第１，第２基準点の座標から上記２点間を結ぶ基準直線を設定する。
【００８８】
ステップＳ２０６でくぼみの測定点を設定する。ステップＳ２０７でサブルーチン３次元座標解析処理を実行し、上記測定点の座標を求める。ステップＳ２０８で上記基準点を結ぶ直線と測定点の垂線の長さｈをくぼみ量として演算し、本ルーチンを終了する。
【００８９】
なお、上記のサブルーチンの処理で被計測物が図１４に示すタービンブレード４３であった場合、第１，第２基準点ＰＴ７，ＰＴ８を結ぶ基準直線とくぼみの測定点ＰＴ９までの距離ｈを求めることになる。
【００９０】
次に、上記ステップＳ１７５，Ｓ１９２，Ｓ２０２、その他のステップでコールされるサブルーチンの３次元座標解析処理について、図１５（Ａ）のフローチャートにより説明する。
【００９１】
ステップＳ２１１にてサブルーチンのパターンマッチング処理を実行して、左右の２画像（ステレオ画像）の対応点であるマッチングポイントを検出する。ステップＳ２１２にて対応点の座標から左右の２画像のずれ量を求める。ステップＳ２１３にて対象としている点の３次元座標を計算し、本ルーチンを終了する。ここで、上記サブルーチンにおける３次元座標解析処理の基本原理について、図１６を用いて説明する。
【００９２】
図１６は、ｘ，ｙ，ｚ　軸をもつ３次元座標系上の右左２画像位置関係を示す図であって、被写体の点Ｐが撮像素子２８の右結像面２８Ｒ、左結像面２８Ｌ上に結像した状態を示している。本図１６において、点ＯＲ，ＯＬを光学系の主点とし、距離ｆを焦点距離とし、点ＱＲ，ＱＬを点Ｐの結像位置とし、さらに、距離Ｌを点ＯＲ−点ＯＬ間の距離とする。
【００９３】
図１６において、直線ＱＲ−ＯＲから次式が成立する。すなわち、
ｘ／ｘ_Ｒ’＝｛ｙ−（Ｌ／２）｝／｛ｙ_Ｒ’−（Ｌ／２）｝＝ｚ／（−ｆ）…（６）
また、直線ＱＬ−ＯＬから次式が成立する。すなわち、
ｘ／ｘ_Ｌ’＝｛ｙ＋（Ｌ／２）｝／｛ｙ_Ｌ’＋（Ｌ／２）｝＝ｚ／（−ｆ）…（７）
この式をｘ、ｙ、ｚ　について解けば、点Ｐの３次元座標が得られる。
【００９４】
なお、実際は像伝送光学系２７の効果により、左右２つの像の光線は折り曲げられて右結像面２８Ｒと左結像面２８Ｌの間隔はもっと小さくなるが、ここでは図を簡略にするために像伝送光学系２７の効果を省いて図示している。
【００９５】
次に、図１５（Ａ）の３次元座標解析処理のステップＳ２１１にてコールされるサブルーチンのパターンマッチング処理について、図１５（Ｂ）のフローチャートにより説明する。
【００９６】
本ルーチンでは、２画像間の対応点を検出するパターンマッチングを行う。図１７は、左右２画面６５，６６の画像をＬＣＤ５上に表示したステレオ計測画面６４を示しており、左画面６５の画像ポイントＰＰ１が右画面６６の画像ポイントＰＰ２に対応していることを示している。
【００９７】
まず、ステップＳ２２１，２２２にて、パターンマッチングを行うパターンの大きさの設定するパターンエリアの絞り込みを行う。本実施の形態の例では、値ｋに対応したパターンエリアを設定する。すなわち、
ｋ＝１ではパターンエリアを３５×３５（ピクセル）、
ｋ＝２ではパターンエリアを２３×２３（ピクセル）、
ｋ＝３ではパターンエリアを１１×１１（ピクセル）、
とし、値ｋを小から大へ切り換えて領域を大から小へ絞り込んでいき、対応点検出の精度を上げるようにする。
【００９８】
ステップＳ２２３にて検索範囲を設定する。すなわち、パターンを探す右画像の領域を決定する。その検索範囲の設定には、エピポーララインに誤差を考慮してエピポーラ±５ピクセル以内とする場合と、モニタ画面上で水平に±７ピクセル以内とする場合と、画面上で手動により指示された略マッチング点を中心に±１０ピクセル以内とする場合がある。なお、上記±１０ピクセルは、手動による誤差を考慮した最適な値である。
【００９９】
ステップＳ２２４〜Ｓ２２６において、設定された検索範囲でのパターンマッチングを行う。このパターンマッチングは、正規化相互相関による対応点検出を行い、最も正規化相互相関係数の大きな座標（Ｘ、Ｙ）を上記対応点とする。
【０１００】
ステップＳ２２７において、値ｋをインクリメントし、その値ｋに対応してパターンを絞り込み、対応点検出を行う。
【０１０１】
ステップＳ２２８においてマッチングポイントの設定を行う。そのとき、正規化相互相関の値をモニタ画面上に表示、これをマッチングの信頼性の尺度としてもよい。また、正規化相互相関の値（−１〜１）が所定の値よりも小さい場合は、手動式のマッチング処理に切り換えてもよい。
【０１０２】
上記パターンマッチングに利用する正規化相互相関関数Ｍ（ｕ，ｖ）は、一般的に以下の式を用いる。すなわち、ｔ（ｘ，ｙ）をテンプレートとし、ｇ（ｘ，ｙ）を画像データとし、ｔ’をテンプレートの平均輝度とし、さらに、ｇ’を画像の平均輝度として、
Ｍ（ｕ，ｖ）＝｛ΣΣ_ｓ（ｇ（ｘ＋ｕ，ｙ＋ｖ）−ｇ’）（ｔ（ｘ，ｙ）−ｔ’）｝
／｛ΣΣ_ｓ（ｇ（ｘ＋ｕ，ｙ＋ｖ）ーｇ’）^２×ΣΣ_ｓ（ｔ（ｘ，ｙ）−ｔ’）^２｝^１／２　　（８）
が適用される。ここでΣΣ_ｓは画素の和をとることを表す。
【０１０３】
本実施の形態による計測例として、例えば補正した画像上で計測を行う。この場合には、リモートコントローラ４で計測実行スイッチ４ｅが操作された場合には、補正された左右の画像がＬＣＤ５に表示される。
【０１０４】
さらにリモートコントローラ４のレバースイッチ４ｂを操作して左右の画像上で、それぞれ計測しようとする点の指示入力を行うことにより、ステレオ計測を行うことができる。
【０１０５】
また、補正した画像を作成して立体視を行うこともできる。この場合にはリモートコントローラ４の計測実行スイッチ４ｅが操作された場合には、補正された左右の画像がＬＣＤ５に表示され、また補正された左右の画像の映像信号がＦＭＤアダプタ６ａに入力され、左右の映像信号に切り離す処理がされた後、ＦＭＤ６に入力される。
【０１０６】
ユーザはＦＭＤ６を装着して、立体視することにより、左右でそれぞれ歪みが補正された良好な画質の左右画像を観察することができ、補正されない左右画像を観察する場合よりもユーザに負担をかけることなく、立体視ができ、ユーザの疲労を大幅に軽減することができる。
【０１０７】
上述の実施の形態の画像データの処理は、簡潔に説明するために画像にモノクロ画像を用いた場合として説明したが、これらの処理は、Ｒ、Ｇ、Ｂ信号によるカラー画像に拡張しても同様に有効であることは勿論である。
【０１０８】
以上述べたように本実施の形態によれば、内視鏡２の先端部２１に１対の対物光学系と、該対物光学系による光学像を伝送する像伝送光学系２７とからなる撮像光学系を設けて撮像素子２８に結像するようにしているので、挿入部２０の外径を細くできると共に、１対の対物光学系により発生する第１の歪みと像伝送光学系２７により発生する第２の歪みとが重なった多重歪みを補正する多重歪み補正手段を設けているので、精度の良い計測ができると共に、多重歪みを補正することにより立体視した場合の疲労を軽減することができる。
【０１０９】
なお、上述の説明では、（ａ１）〜（ｄ）の光学データを作成する場合、計測内視鏡装置１にパソコン３１を接続してパソコンタ３１の動作により行うようにしていたが、第１の実施の形態の変形例として、パソコン３１を接続することなく、計測内視鏡装置１自体で行うようにしても良い。
【０１１０】
この場合には、図２に示す計測処理部１８における例えばＣＰＵ１８ａは、メモリカード３３に記録されている光学データに基づいてステレオ計測を行う処理の他に、（１−１）〜（１−１２）の多重歪み補正の処理、つまり多重歪み補正処理１８ｄの機能等を含む処理を行う。ＣＰＵ１８ａによるこの多重歪み補正処理１８ｄの機能を図２では２点鎖線で示している。
【０１１１】
（第２の実施の形態）
次に本発明の第２の実施の形態を図１８ないし図２５を参照して説明する。本実施の形態の計測内視鏡装置は、図１の計測内視鏡装置１において、内視鏡２の先端部２１の構成を変更したものである。
【０１１２】
つまり、内視鏡２の先端部２１は図１８に示すように先端部本体３９と、この先端部本体３９に着脱自在の光学アダプタ４６とから構成されるようになっている。
【０１１３】
光学アダプタ４６には２つの照明窓２４と、観察窓２５とが設けてある。図１８のＢ−Ｂ断面による先端部２１の構成を図１９に示す。
【０１１４】
図１９に示すように先端部本体３９には撮像素子２８が取り付けてあり、この撮像素子２８の前面側はカバーガラス４７で保護され、このカバーガラス４７は光学アダプタ４６側のカバーガラス４８と対向する。
【０１１５】
この光学アダプタ４６の後端側の外周面には、固定リング４９が設けてあり、先端部本体３９の外周面に設けた雄ネジ部にこの固定リング４９の後端内視鏡内周面に設けた雌ネジ部を螺合させることにより、着脱自在に取り付けられるようにしている。
【０１１６】
なお、先端部本体３９の先端面の外周面には位置決め用の凹部が、光学アダプタ３９側には位置決め用ピンが設けてあり、光学アダプタ３９を取り付ける際に凹部とピンとにより周方向の位置決めがされる。この光学アダプタ４６側には、対物レンズ２６Ｒ、２６Ｌと像伝送光学系２７とが取り付けられている。
【０１１７】
その他は基本的には第１の実施の形態と同様の構成である。
【０１１８】
次に本実施の形態の作用を説明する。
【０１１９】
本計測内視鏡装置に適用される光学アダプタ４６は、生産工程にて個体の異なる光学アダプタ毎に図２０に示す生産測定治具５１を構成する撮像素子内蔵のマスタ撮像ユニット５２に取り付けて、次の（ａ１）〜（ｄ）に示す各光学アダプタ４６の特有の光学データが測定される。その光学データを記録媒体である例えば、メモリカード３３に記録する。このメモリカード３３に記録された特有の光学データは、光学アダプタ４６の特性と、一対一で対応することになって、出荷後、１つの組み合わせのものとして扱われる。
【０１２０】
上述の特有の光学データは、
（ａ１）２つの対物光学系の幾何学的歪み補正テーブル
（ａ２）像伝送光学系の幾何学歪み補正テーブル
（ｂ）左右の結像光学系それぞれの焦点距離
（ｃ）左右の結像光学系の主点間の距離
（ｄ）左右の結像光学系それぞれの画像上での光軸位置座標
（ｅ）左右の結像光学系それぞれの画像がマスタの撮像素子上に結像するときの位置情報である。
【０１２１】
本実施の形態は、（ｅ）の項目が加わったことが第１の実施の形態と異なる点であり、その他は第１の実施の形態と同様である。
【０１２２】
上記特有の光学データ採りを行った後の光学アダプタ４６は、内視鏡２に取り付け、計測内視鏡装置において、次に示す（１）〜（８）の処理を行って各種寸法計測を行うことができる。
【０１２３】
すなわち、
（１）上記メモリカード３３から上記（ａ１）〜（ｅ）の光学データを読み込む。（２）本計測内視鏡装置を用いて白い被写体を撮像する。
【０１２４】
（３）上記（ｅ）のデータおよび上記（２）の撮像データを用いて本光学アダプタ４６と内視鏡２との組み合わせによる画像位置のずれを求める。
【０１２５】
（４）上記（３）のデータおよび上記（１）のデータを用いて、本内視鏡２に対する幾何学的歪み補正を行う変換テーブルを作成する。
【０１２６】
（５）本内視鏡２にて被写体である被計測物を撮像し、画像を取り込む。
【０１２７】
（６）上記の取り込んだ画像を上記（３）で作成したテーブルをもとに座標変換する。
【０１２８】
（７）座標変換された画像を基に、上記（２）の撮像データのマッチングにより任意の点の３次元座標を求める。
【０１２９】
（８）上記３次元座標を基に各種寸法計測を行う。
【０１３０】
図２０は、前述した光学アダプタ４６に特有の光学データを測定するための生産測定治具５１による測定状態を示す斜視図である。
【０１３１】
上記生産測定治具５１は、光学アダプタ４６が装着可能であって、内視鏡２の先端部本体３９と同様の構造を有するマスタ撮像ユニット５２と、このマスタ撮像ユニット５２からの信号線が接続されるＣＣＵ５３と、メモリカード３３が着脱可能なメモリカードスロット５４を有し、上記ＣＣＵ５３からの画像データに対する画像処理を行うパソコン３１、および光学アダプタ４６の光学特性を解析するためのチャート５４とで構成されている。
【０１３２】
上記生産測定治具５１による光学データの取り込みを行う場合、まず、図２０に示すように、光学アダプタ４６をマスタ撮像ユニット５２に取り付け、チャート５４の像を光学アダプタ４６を介して取り込み、その画像データに基づいてパーソナルコンピュータ３１にて、画像処理を行い、前記（ａ１）〜（ｅ）の光学データを求め、メモリカード３３に記録する。
【０１３３】
次に、上記光学アダプタ４６に特有の各光学データを具体的に説明する。
【０１３４】
（ａ１）（ａ２）幾何学的歪み補正テーブルについて、
一般にレンズ系による画像には光学的な歪みがある。計測を行う場合にはこの歪みが大きな誤差原因となるため、座標変換を行うことによりこの歪みを取り除くことができる。座標変換は、光軸中心を中心にして行ってもよいし、より正確に補正する場合は、光学系の幾何学的歪みの中心を用いるとよい。また、２つの画像の幾何学的歪み補正テーブルは、右画像、左画像別々に設けてもよいし、２つをまとめて１つのテーブルにしてもよい。この補正は第１の実施の形態と同様に行うことができる。
【０１３５】
また、（ｂ）、（ｃ）、（ｄ）は第１の実施の形態と同様であるのでその説明を省略する。
【０１３６】
（ｅ）２つの画像がマスタの撮像素子上に結像するときの位置情報に関しては、基準ラインＶ（垂直ライン）の輝度データとして、
ＰＶ（１００，Ｙｎ）　但し、Ｙｎ＝１，２，３，…４８０
および、基準ラインＨ（水平ライン）の輝度データとして、
ＰＨ（Ｘｎ，１００）　但し、Ｘｎ＝１，２，３，…６４０
を測定して記録する。
【０１３７】
次に計測動作としてのステレオ計測処理について、図２１，２２のフローチャート、および図２３〜図２５等を用いて説明する。
【０１３８】
本ステレオ計測処理は、光学アダプタ４６の２つの対物レンズ２６Ｒ、対物レンズ２６Ｌにより取り込まれる視差のある右左の画像データに基づいて、３次元座標の計測を行う処理である。
【０１３９】
まず、図２１のステップＳ３０において、初期化を行うか否かをチェックする。すなわち、本計測ソフト立ち上げて最初の状態の場合と、光学アダプタ４６を交換したか、内視鏡を交換した場合には、ステップＳ３１〜３５のルーチンに進み、初期化を実行する。それ以外の場合は、ステップＳ１１に進み、すでに設定されている補正画像情報データをオリジナルデータとして読み込み、ステップＳ１２に進む。
【０１４０】
上記ステップＳ３１においては、補正画像データの読み込みが行われる。すなわち、光学アダプタ４６と一対一で対応した前記（ａ１）〜（ｅ）の特有の光学データをメモリカード３３から読み込み、これをマスタデータとする。
【０１４１】
続いて、ステップＳ３２では、白い被写体のイメージを取り込む。すなわち、光学アダプタ４６を取り付けた計測内視鏡装置により白い被写体を写し出し、その画像を映像信号処理回路１２を介して計測処理部１８のＣＰＵ１８ａはＲＡＭ１８ｃ上に取り込む。
【０１４２】
続いて、ステップＳ３３では、後述するサブルーチン画像位置設定処理（図２２（Ａ）参照）を実行する。このサブルーチンの処理は、光学アダプタ４６の生産時に光学データを取得したマスタ撮像ユニット５２と内視鏡２との特性の違いにより生じる画像の座標のずれ量を求める処理である。
【０１４３】
ステップＳ３４では、テーブルの設定を行う。この処理では、ステップＳ３１で取り込んだ前記光学データのうち（ａ１）、（ａ２）の２種類の画像の幾何学的歪み補正テーブルをステップＳ３３で求めた座標のずれ量に応じて補正した補正テーブルが作成される。すなわち、以下の２つで構成されるオリジナルデータが作成される。
【０１４４】
ウエイトテーブルとして
Ｗ１’（Ｘ，Ｙ）＝Ｗ１（Ｘ＋ΔＸ，Ｙ＋ΔＹ）
Ｗ２’（Ｘ，Ｙ）＝Ｗ２（Ｘ＋ΔＸ，Ｙ＋ΔＹ）
Ｗ３’（Ｘ，Ｙ）＝Ｗ３（Ｘ＋ΔＸ，Ｙ＋ΔＹ）
Ｗ４’（Ｘ，Ｙ）＝Ｗ４（Ｘ＋ΔＸ，Ｙ＋ΔＹ）…（９）
また、座標参照テーブルとして、
ＱＸ’（Ｘ，Ｙ）＝ＱＸ（Ｘ，Ｙ）＋ΔＸ
ＱＹ’（Ｘ，Ｙ）＝ＱＹ（Ｘ，Ｙ）＋ΔＹ　　　…（１０）
但し、「’付き」は、補正後を示し、「’なし」は、補正前を示す。また、ΔＸ，ΔＹは、マスタで測定した画像と光学アダプタ４６を本装置に装着したときの画像のずれ量を示す。
【０１４５】
ステップＳ３５では、補正画像情報データの保存を行う。すなわち、ステップＳ３４で求めたオリジナルデータをメモリカード３３に保存する。
【０１４６】
そこで、ステップＳ１２において、イメージデータを取り込む。すなわち、計測すべき物体を撮像し、メモリ上に原画像データとして取り込む。そして、ステップＳ１３で後述するサブルーチンの画像補正処理（図２２（Ｂ）参照）を実行し、原画像の画像補正を行う。すなわち、ステップＳ１２で取り込んだ原画像データをステップＳ３４で作成したオリジナルテーブルの基づいて座標変換し、計測を行うための計測画像データを得る。
【０１４７】
続いて、ステップＳ１４にて上記計測画像データに基づき、上述したサブルーチンの計測処理（図１０参照）を実行し、３次元座標の演算等を行って選択された計測を行い、本ステレオ計測処理を終了する。
【０１４８】
ここで、上記ステップＳ３３においてコールされるサブルーチンの画像位置設定処理について、図２２（Ａ）のフローチャートを用いて説明する。
【０１４９】
まず、ステップＳ１５２において、視野形状パターンを呼び出す。すなわち、上記ステップＳ３１で取り込んだ特有の光学データのうち（ｅ）のマスタの撮像素子に対する結像の位置情報を示す輝度データとして、
ＰＶ（１００，Ｙｎ），但し、Ｙｎ＝１，２，３，…４８０
ＰＨ（Ｘｎ，１００），但し、Ｘｎ＝１，２，３，…６４０
を呼び出す。図２３は、光学アダプタ４６をマスタ撮像ユニット５２に装着した状態で白色チャート５４を撮像した画像を示す図であり、垂直ラインＰＶ，水平ラインＰＨがそれぞれ上記輝度データを与えるラインである。
【０１５０】
ステップＳ１５３にてパターンマッチングおよび視野位置設定、光軸位置設定を行う。図２４は、光学アダプタ４６を装着した計測内視鏡装置により白色チャート５４の画像を取り込んだ画像を示す図である。結像の位置情報を示す輝度データとして、
ＰＶ’（１００，Ｙｎ），但し、Ｙｎ＝１，２，３，…４８０
ＰＨ’（Ｘｎ，１００），但し、Ｘｎ＝１，２，３，…６４０
で与えられる。
【０１５１】
また、図２５は、上記マスタ、および、計測内視鏡装置１に対する視野形状パターンを与える輝度とを縦軸とし、横軸を画像のアドレスとした輝度の変化を示す図であって、図２５（Ａ）がラインＰＶ、および、ラインＰＶ’の輝度変化を示し、図２５（Ｂ）がラインＰＨ、および、ラインＰＨ’の輝度変化を示している。
【０１５２】
上記ステップＳ１５３のパターンマッチング処理では、図２５（Ａ）、および、図２５（Ｂ）の輝度変化に基づき、正規化相互相関を用いてマスタに対する画像の位置ずれΔＸ、ΔＹを求める。さらに、計測系光軸位置は、右左光学系に関して水平，垂直座標ｘＲ’，ｙＲ’とｘＬ’，ｙＬ’は、次の式で求められる。すなわち、
ｘＲ’＝ＸＲ＋ΔＸ
ｙＲ’＝ＹＲ＋ΔＹ
ｘＬ’＝ＸＬ＋ΔＸ
ｙＬ’＝ＹＬ＋ΔＹ　…（１１）
となる。
【０１５３】
また、上記ステップＳ１３（図２１参照）において、コールされるサブルーチンの画像補正処理について、図２２（Ｂ）のフローチャートを用いて説明する。本サブルーチンの処理により計測画像上の画素データＰ（Ｘ，Ｙ）を得ることができる。まず、ステップＳ１６２において、前記ステップＳ３４（図２１参照）にて作成した補正テーブルであるオリジナルデータを取り込む。ステップＳ１６３では、補正した画像の座標、すなわち、計測画像の座標をＸ，Ｙと設定する。
【０１５４】
ステップＳ１６４において、補正座標Ｘ，Ｙを得るために必要な原画像の座標を参照座標とし、前記ステップＳ３４で作成したＱＸ’（Ｘ，Ｙ）、ＱＹ’（Ｘ，Ｙ）により補正座標Ｘ，Ｙを得る。そして、ステップＳ１６５にて、前記ステップＳ３４で作成したウエイトテーブルを取り込み、ステップＳ１６６で補正後の座標上の画素データＰ（Ｘ，Ｙ）を次式より求める。すなわち、

により求められる。
【０１５５】
続いて、ステップＳ１６７において、補正画像範囲を設定し、本ルーチンを終了する。なお、このサブルーチンの画像補正処理は、左右２画像に対して行い、原画像を計測画像に変換する。
【０１５６】
図２１のステップＳ１４にてコールされるサブルーチンの計測処理は図１０で説明したので省略する。
【０１５７】
また、本実施の形態の計測例として、補正は輝度情報に対してのみ内部で行い画像は作成せず、補正前の画像上で計測点を指定するようにしても良い。
【０１５８】
この場合には、補正画像作成処理を簡略化できるので、画像取り込みから計測開始までの時間を短縮できる。
【０１５９】
本実施の形態の使用例としては、第１の実施の形態で説明したように行うことができると共に、以下のように補正は輝度情報に対してのみ内部で行い、その場合には補正された画像を作成しないで、補正前の画像上で計測点を指定するようにしても良い。
【０１６０】
この使用例では、補正画像の作成処理を簡略化できるので、画像取り込みから計測開始までの時間を短縮することができる。
【０１６１】
なお、本実施の形態では、ステレオ式の光学アダプタ４６を装着した場合で説明したが、他の光学アダプタを装着して使用することもできる。
【０１６２】
以上、説明したように本実施の形態の計測内視鏡装置によれば、計測精度を向上するのために、光学アダプタ４６に一対一で対応するように特有の光学データを記録媒体のメモリカード３３で提供し、生産測定治具５１（マスタ撮像ユニット５２）と内視鏡画像との座標のずれを補正したことにより、計測精度を向上させることができる。
【０１６３】
また、計測用の光学アダプタ４６の生産に生産測定治具５１（マスタ撮像装置５２）を用いて特有の光学データ取りを行ったため、生産性が向上して計測内視鏡装置を低価格で提供できるようになる。
【０１６４】
また、先端部２１に計測用の光学アダプタ４６を着脱自在に設けたため、他の計測以外の光学アダプタとの交換が可能になり、内視鏡の汎用性を向上させ、ユーザの金銭的負担を抑えることができる。
【０１６５】
また、本実施の形態の例では画像の座標変換に使用するウエイトテーブルＷ１〜Ｗ４、座標参照テーブルＱＸ、ＱＹを生産測定治具５１のコンピュータ３１で作成して記録媒体のメモリカード３３に記録したため、内視鏡装置内部の計測処理部でテーブルを作成する必要がない。したがって、計測処理を短時間で行うことができる。
【０１６６】
さらに、正規化相互相関の値をＬＣＤ５の画面上に表示することにより、操作者は、マッチングの信頼性、あるいは、計測の信頼性を知ることができる。また、正規化相互相関の値が小さい場合は、手動式のマッチングに切り換えることができ、精度の高い計測が可能になる。
【０１６７】
（第３の実施の形態）
次に本発明の第３の実施の形態を図２６を参照して説明する。第２の実施の形態では、対物光学系及び像伝送光学系２７がすべて光学アダプタ４６側にあり、第２歪みはまとめて補正するようにしていたが、本実施の形態では対物光学系と像伝送光学系２７の一部が光学アダプタ４６側にあり、第２歪みはまとめて補正するようにしている。
【０１６８】
本実施の形態における内視鏡２の先端部２１の構成を図２６に示す。先端部２１は先端部本体５９と、この先端部本体５９に着脱自在の光学アダプタ６０とからなり、先端部本体５９には撮像素子２８と像伝送光学系２７の後段側光学系２７ｂとが設けてある。
【０１６９】
また、光学アダプタ６０には２つの対物レンズ２６Ｒ、２６Ｌと、像伝送光学系２７の前段光学系２７ａ、を備え、先端部本体５９に固定リング４９で着脱自在である。この場合における先端部本体５９と光学アダプタ６０とに分割する部分、つまりカバーガラス６１，６２の面で光束径が太くなり、かつ実像が結像しないことでごみが付いても画像への影響が小さいようにしている。
【０１７０】
本実施の形態では式（１−１）〜（１−４）で前段と後段を合わせた像伝送光学系２７の幾何学的歪みを補正し、式（１−５）〜（１−１２）で対物光学系の幾何学的歪みを補正する。
【０１７１】
その他は第２の実施の形態と同様の構成及び作用となる。また、本実施の形態の効果も第２の実施の形態とほぼ同様となる。
【０１７２】
（第４の実施の形態）
次に本発明の第４の実施の形態を図２７を参照して説明する。本実施の形態における内視鏡装置は第３の実施の形態と同様である。
【０１７３】
つまり、内視鏡２の先端部２１は図２６に示す構成であり、像伝送光学系２７の前段側光学系２７ａが光学アダプタ６０側にあり、後段側光学系２７ｂは先端部本体５９側に設けてある。
【０１７４】
しかし、本実施の形態では第２歪みを前段側と後段側とで別々に補正を行うようにしたものである。
【０１７５】
計測処理部１８には幾何学的な多重歪み補正処理の機能が設けてあり、幾何学的な多重歪み補正処理の機能は第１歪み補正処理の機能とと第２歪み前段補正処理の機能と、第２歪み後段補正処理の機能とから構成される。
【０１７６】
以下に示した式（１３−１）〜（１３−４）で伝送光学系後段の幾何学的歪みを補正し、式（１３−５）〜（１３−８）で像伝送光学系前段の幾何学的歪みを補正し、式（１３−９）〜（１３−１６）で対物光学系の幾何学的歪みを補正する。
【０１７７】
ｘ，ｙ：補正前座標
ｘ_Ｒ’，ｙ_Ｒ’：補正後右画面座標
ｘ_Ｌ’，ｙ_Ｌ’：補正後左画面座標
ｃ_Ｂｘ，ｃ_Ｂｙ：第２歪み後段中心座標
ｃ_Ｆｘ，ｃ_Ｆｙ：第２歪み前段中心座標
ｃ_Ｒｘ，ｃ_Ｒｙ：第１歪み右中心座標
ｃ_Ｌｘ，ｃ_Ｌｙ：第１歪み左中心座標
ｋ_ｘ，ｋ_ｙ，ａ_ｉｊ，ｂ_ｉｊ：第２歪み補正係数
ｋ_Ｒｘ，ｋ_Ｒｙ，ａ_Ｒｉｊ，ｂ_Ｒｉｊ：第１歪み右補正係数
ｋ_Ｌｘ，ｋ_Ｌｙ，ａ_Ｌｉｊ，ｂ_Ｌｉｊ：第１歪み左補正係数
第２歪み後段補正：
ｆ_２Ｂｘ（ｘ，ｙ）≡ｋ_Ｂｘ（ａ_Ｂ００＋ａ_Ｂ１２ｘｙ^２＋ａ_Ｂ１４ｘｙ^４＋ａ_Ｂ１６ｘｙ^６＋ａ_Ｂ３０ｘ^３＋ａ_Ｂ３２ｘ^３ｙ^２＋ａ_Ｂ３４ｘ^３ｙ^４
＋ａ_Ｂ５０ｘ^５＋ａ_Ｂ５２ｘ^５ｙ^２＋ａ_Ｂ７０ｘ^７）＋ｃ_Ｂｘ　（１３−１）
ｆ_２Ｂｙ（ｘ，ｙ）≡ｋ_Ｂｙ（ｂ_Ｂ００＋ｂ_Ｂ２１ｘ^２ｙ＋ｂ_Ｂ４１ｘ^４ｙ＋ｂ_Ｂ６１ｘ^６ｙ＋ｂ_Ｂ０３ｙ^３＋ｂ_Ｂ２３ｘ^２ｙ^３＋ｂ_Ｂ４３ｘ^４ｙ^３
＋ｂ_Ｂ０５ｙ^５＋ｂ_Ｂ２５ｘ^２ｙ^５＋ｂ_Ｂ０７ｙ^７）＋ｃ_Ｂｙ　（１３−２）
ｘ_２１＝ｆ_２ｘ（ｘ−ｃ_Ｂｘ，ｙ−ｃ_Ｂｙ）　（１３−３）
ｙ_２１＝ｆ_２ｙ（ｘ−ｃ_Ｂｘ，ｙ−ｃ_Ｂｙ）　（１３−４）
第２歪み前段補正：
ｆ_２Ｆｘ（ｘ，ｙ）≡ｋ_Ｆｘ（ａ_Ｆ００＋ａ_Ｆ１２ｘｙ^２＋ａ_Ｆ１４ｘｙ^４＋ａ_Ｆ１６ｘｙ^６＋ａ_Ｆ３０ｘ^３＋ａ_Ｆ３２ｘ^３ｙ^２＋ａ_Ｆ３４ｘ^３ｙ^４
＋ａ_Ｆ５０ｘ^５＋ａ_Ｆ５２ｘ^５ｙ^２＋ａ_Ｆ７０ｘ^７）＋ｃ_Ｆｘ　（１３−５）
ｆ_２Ｆｙ（ｘ，ｙ）≡ｋ_Ｆｙ（ｂ_Ｆ００＋ｂ_Ｆ２１ｘ^２ｙ＋ｂ_Ｆ４１ｘ^４ｙ＋ｂ_Ｆ６１ｘ^６ｙ＋ｂ_Ｆ０３ｙ^３＋ｂ_Ｆ２３ｘ^２ｙ^３＋ｂ_Ｆ４３ｘ^４ｙ^３
＋ｂ_Ｆ０５ｙ^５＋ｂ_Ｆ２５ｘ^２ｙ^５＋ｂ_Ｆ０７ｙ^７）＋ｃ_Ｆｙ　（１３−６）
ｘ_２＝ｆ_２ｘ（ｘ_２１−ｃ_Ｆｘ，ｙ_２１−ｃ_Ｆｙ）　（１３−７）
ｙ_２＝ｆ_２ｙ（ｘ_２１−ｃ_Ｆｘ，ｙ_２１−ｃ_Ｆｙ）　（１３−８）
第１歪み右補正：
ｆ_Ｒｘ（ｘ，ｙ）≡ｋ_Ｒｘ（ａ_Ｒ００＋ａ_Ｒ１２ｘｙ^２＋ａ_Ｒ１４ｘｙ^４＋ａ_Ｒ１６ｘｙ^６＋ａ_Ｒ３０ｘ^３＋ａ_Ｒ３２ｘ^３ｙ^２＋ａ_Ｒ３４ｘ^３ｙ^４
＋ａ_Ｒ５０ｘ^５＋ａ_Ｒ５２ｘ^５ｙ^２＋ａ_Ｒ７０ｘ^７）＋ｃ_Ｒｘ　（１３−９）
ｆ_Ｒｙ（ｘ，ｙ）≡ｋ_Ｒｙ（ｂ_Ｒ００＋ｂ_Ｒ２１ｘ^２ｙ＋ｂ_Ｒ４１ｘ^４ｙ＋ｂ_Ｒ６１ｘ^６ｙ＋ｂ_Ｒ０３ｙ^３＋ｂ_Ｒ２３ｘ^２ｙ^３＋ｂ_Ｒ４３ｘ^４ｙ^３
＋ｂ_Ｒ０５ｙ^５＋ｂ_Ｒ２５ｘ^２ｙ^５＋ｂ_Ｒ０７ｙ^７）＋ｃ_Ｒｙ　（１３−１０）
Ｘ_Ｒ＝ｆ_Ｒｘ（ｘ_２−ｃ_Ｒｘ，ｙ_２−ｃ_Ｒｙ）　（１３−１１）
Ｙ_Ｒ＝ｆ_Ｒｙ（ｘ_２−ｃ_Ｒｘ，ｙ_２−ｃ_Ｒｙ）　（１３−１２）
第１歪み左補正：
ｆ_Ｌｘ（ｘ，ｙ）≡ｋ_Ｌｘ（ａ_Ｌ００＋ａ_Ｌ１２ｘｙ^２＋ａ_Ｌ１４ｘｙ^４＋ａ_Ｌ１６ｘｙ^６＋ａ_Ｌ３０ｘ^３＋ａ_Ｌ３２ｘ^３ｙ^２＋ａ_Ｌ３４ｘ^３ｙ^４
＋ａ_Ｌ５０ｘ^５＋ａ_Ｌ５２ｘ^５ｙ^２＋ａ_Ｌ７０ｘ^７）＋ｃ_Ｌｘ　（１３−１３）
ｆ_Ｌｙ（ｘ，ｙ）≡ｋ_Ｌｙ（ｂ_Ｌ００＋ｂ_Ｌ２１ｘ^２ｙ＋ｂ_Ｌ４１ｘ^４ｙ＋ｂ_Ｌ６１ｘ^６ｙ＋ｂ_Ｌ０３ｙ^３＋ｂ_Ｌ２３ｘ^２ｙ^３＋ｂ_Ｌ４３ｘ^４ｙ^３
＋ｂ_Ｌ０５ｙ^５＋ｂ_Ｌ２５ｘ^２ｙ^５＋ｂ_Ｌ０７ｙ^７）＋ｃ_Ｌｙ　（１３−１４）
ｘ_Ｌ’＝ｆ_Ｌｘ（ｘ_２−ｃ_Ｌｘ，ｙ_２−ｃ_Ｌｙ）　（１３−１５）
ｙ_Ｌ’＝ｆ_Ｌｙ（ｘ_２−ｃ_Ｌｘ，ｙ_２−ｃ_Ｌｙ）　（１３−１６）
光学アダプタ５９を実際に装着される内視鏡２に応じて、第２歪み後段中心と第２歪み前段中心の位置のずれ量が変化した場合にも対応でき、より高精度な補正を行えるようにしている。
【０１７８】
この場合の光学データは、
（ａ１）２つの対物光学系の幾何学的歪み補正テーブル
（ａ２）像伝送光学系前段の幾何学歪み補正テーブル
（ａ３）像伝送光学系後段の幾何学歪み補正テーブル
（ｂ）左右の結像光学系それぞれの焦点距離
（ｃ）左右の結像光学系の主点間の距離
（ｄ）左右の結像光学系それぞれの画像上での光軸位置座標
（ｅ）左右の結像光学系それぞれの画像がマスタの撮像素子上に結像するときの位置情報である。
【０１７９】
上記光学データを参照して、図２７に示すように幾何学的な多重歪み補正を行う。
【０１８０】
つまり、図２７に示すように座標変換が開始すると、最初のステップＳ４１で第２歪み後段補正テーブルから第２歪み後段補正データを読み出し、次のステップＳ４２で像伝送光学系２７の後段側光学系２７ｂの第２歪み後段補正を行う。次のステップＳ４３で第２歪み前段補正テーブルから第２歪み前段補正データを読み出し、次のステップＳ４４で像伝送光学系２７の前段側光学系２７ａの第２歪み後段補正を行う。
【０１８１】
次のステップＳ４５で第１歪み補正テーブルから第１歪み補正データを読み出し、次のステップＳ４６で対物光学系の第１歪み補正を行う。
【０１８２】
このように補正を行うことにより、光学アダプタ５９を実際に装着される内視鏡２に応じて、より高精度な補正を行える。その他は第３の実施の形態と同様の効果を有する。
【０１８３】
（第５の実施の形態）
次に本発明の第５の実施の形態を説明する。本実施の形態は第１〜第３の実施の形態において、式（１−１）〜（１−４）の第２歪み補正式の代わりに以下の式（１４−１）〜（１４−５）を採用したものである。この場合の効果は第１〜第３の実施の形態とほぼ同様のものとなる。
【０１８４】
ｈ_２（ｘ，ｙ）≡Ｋ_２　ｓｉｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_２Ｋ_２）｝　（１４−１）
ｈ_２（ｘ，ｙ）≡Ｋ_２　ｔａｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_２Ｋ_２）｝　（１４−１’）
（１４−１）と（１４−１’）はレンズの幾何学的歪みの特性により使い分ける。
【０１８５】
ｆ_２ｘ（ｘ，ｙ）≡ｋ_２ｘｄ_２　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_２（ｘ，ｙ）｝＋ｃ_ｘ　（１４−３）
ｆ_２ｙ（ｘ，ｙ）≡ｋ_２ｙｄ_２　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_２（ｘ，ｙ）｝＋ｃ_ｙ　（１４−３）
ｘ_２＝ｆ_２ｘ（ｘ−ｃ_ｘ，ｙ−ｃ_ｙ）　（１４−４）
ｙ_２＝ｆ_２ｙ（ｘ−ｃ_ｘ，ｙ−ｃ_ｙ）　（１４−５）
ここで、
ｘ，ｙ：補正前座標
ｘ_２，ｙ_２：第２歪み補正後座標
ｃ_ｘ，ｃ_ｙ：第２歪み中心座標
Ｋ_２，ｄ_２，ｋ_２ｘ，ｋ_２ｙ：第２歪み補正係数
である。
【０１８６】
また、第４の実施の形態に対しては、式（１３−１）〜（１３−８）の第２歪み補正式の代わりに以下の式（１５−１）〜（１５−１０）を採用すれば良い。この場合の効果は第４の実施の形態とほぼ同様のものとなる。
【０１８７】
第２歪み後段補正：
ｈ_Ｂ（ｘ，ｙ）≡Ｋ_Ｂ　ｓｉｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＢＫ_Ｂ）｝　（１５−１）
ｈ_Ｂ（ｘ，ｙ）≡Ｋ_Ｂ　ｔａｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＢＫ_Ｂ）｝　（１５−１’）
（１５−１）と（１５−１’）はレンズの幾何学的歪みの特性により使い分ける。
【０１８８】
ｆ_Ｂｘ（ｘ，ｙ）≡ｋ_Ｂｘｄ_Ｂ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｂ（ｘ，ｙ）｝＋ｃ_Ｂｘ　（１５−２）
ｆ_Ｂｙ（ｘ，ｙ）≡ｋ_Ｂｙｄ_Ｂ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｂ（ｘ，ｙ）｝＋ｃ_Ｂｙ　（１５−３）
ｘ_Ｂ＝ｆ_Ｂｘ（ｘ−ｃ_Ｂｘ，ｙ−ｃ_Ｂｙ）　（１５−４）
ｙ_Ｂ＝ｆ_Ｂｙ（ｘ−ｃ_Ｂｘ，ｙ−ｃ_Ｂｙ）　（１５−５）
第２歪み前段補正：
ｈ_Ｆ（ｘ，ｙ）≡Ｋ_Ｆ　ｓｉｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＦＫ_Ｆ）｝　（１５−６）
ｈ_Ｆ（ｘ，ｙ）≡Ｋ_Ｆ　ｔａｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＦＫ_Ｆ）｝　（１５−６’）
（１５−６）と（１５−６’）はレンズの幾何学的歪みの特性により使い分ける。
【０１８９】
ｆ_Ｆｘ（ｘ，ｙ）≡ｋ_Ｆｘｄ_Ｆ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｆ（ｘ，ｙ）｝＋ｃ_Ｆｘ　（１５−７）
ｆ_Ｆｙ（ｘ，ｙ）≡ｋ_Ｆｙｄ_Ｆ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｆ（ｘ，ｙ）｝＋ｃ_Ｆｙ　（１５−８）
ｘ_２＝ｆ_Ｆｘ（ｘ−ｃ_Ｆｘ，ｙ−ｃ_Ｆｙ）　（１５−９）
ｙ_２＝ｆ_Ｆｙ（ｘ−ｃ_Ｆｘ，ｙ−ｃ_Ｆｙ）　（１５−１０）
ここで、
ｘ，ｙ：補正前座標
ｘ_２，ｙ_２：第２歪み補正後座標
ｃ_Ｆｘ，ｃ_Ｆｙ：第２歪み前段中心座標
ｃ_Ｂｘ，ｃ_Ｂｙ：第２歪み後段中心座標
Ｋ_Ｆ，ｄ_Ｆ，ｋ_Ｆｘ，ｋ_Ｆｙ：第２歪み前段中心座標
Ｋ_Ｂ，ｄ_Ｂ，ｋ_Ｂｘ，ｋ_Ｂｙ：第２歪み後段中心座標
また、上述した第１ないし第４の実施の形態において、第１歪み補正の座標変換式を以下のようにしても良い。
【０１９０】
第１歪み右補正：
ｈ_Ｒ（ｘ，ｙ）≡Ｋ_Ｒ　ｓｉｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＲＫ_Ｒ）｝　（１６−１）
ｈ_Ｒ（ｘ，ｙ）≡Ｋ_Ｒ　ｔａｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＲＫ_Ｒ）｝　（１６−１’）
（１６−１）と（１６−１’）はレンズの幾何学的歪みの特性により使い分ける。
【０１９１】
ｆ_Ｒｘ（ｘ，ｙ）≡ｋ_Ｒｄ_Ｒ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｒ（ｘ，ｙ）｝＋ｃ_Ｒｘ　（１６−２）
ｆ_Ｒｙ（ｘ，ｙ）≡ｋ_Ｒｄ_Ｒ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｒ（ｘ，ｙ）｝＋ｃ_Ｒｙ　（１６−３）
ｘ_Ｒ＝ｆ_Ｒｘ（ｘ−ｃ_Ｒｘ，ｙ−ｃ_Ｒｙ）　（１６−４）
ｙ_Ｒ＝ｆ_Ｒｙ（ｘ−ｃ_Ｒｘ，ｙ−ｃ_Ｒｙ）　（１６−５）
第１歪み左補正：
ｈ_Ｌ（ｘ，ｙ）≡Ｋ_Ｌ　ｓｉｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＬＫ_Ｌ）｝　（１６−６）
ｈ_Ｌ（ｘ，ｙ）≡Ｋ_Ｌ　ｔａｎ^−１｛（ｘ^２＋ｙ^２）^１／２／（ｄ_ＬＫ_Ｌ）｝　（１６−６’）
（１６−６）と（１６−６’）はレンズの幾何学的歪みの特性により使い分ける。
【０１９２】
ｆ_Ｌｘ（ｘ，ｙ）≡ｋ_Ｌｄ_Ｌ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｌ（ｘ，ｙ）｝＋ｃ_Ｌｘ　（１６−７）
ｆ_Ｌｙ（ｘ，ｙ）≡ｋ_Ｌｄ_Ｌ　ｃｏｓ｛ｔａｎ^−１（ｙ／ｘ）｝ｔａｎ｛ｈ_Ｌ（ｘ，ｙ）｝＋ｃ_Ｌｙ　（１６−８）
ｘ_Ｌ＝ｆ_Ｌｘ（ｘ_２−ｃ_Ｌｘ，ｙ_２−ｃ_Ｌｙ）　（１６−９）
ｙ_Ｌ＝ｆ_Ｌｙ（ｘ_２−ｃ_Ｌｘ，ｙ_２−ｃ_Ｌｙ）　（１６−１０）
ここで、
ｘ_２，ｙ_２：第２歪み補正後座標
ｘ_Ｒ，ｙ_Ｒ：補正後右画面座標
ｘ_Ｌ，ｙ_Ｌ：補正後左画面座標
ｃ_Ｒｘ，ｃ_Ｒｙ：第１歪み右中心座標
ｃ_Ｌｘ，ｃ_Ｌｙ：第１歪み左中心座標
Ｋ_Ｒ，ｄ_Ｒ，ｋ_Ｒｘ，ｋ_Ｒｙ：第１歪み右補正係数
Ｋ_Ｌ，ｄ_Ｌ，ｋ_Ｌｘ，ｋ_Ｌｙ：第１歪み左補正係数
である。
【０１９３】
なお、上述したように多重歪みを補正する光学データを用いて、１対の対物光学系と像伝送光学系を介して撮像素子に結像する構成を備えた既存の内視鏡に対して、その画像を補正するのに適用することもできる。
【０１９４】
この場合には、例えば光学データを記録したメモリカード３３等の記録媒体を介して画像補正を行う画像処理手段に読み込ませる等して多重歪み補正を行うことにより、精度の高い計測や画像表示が可能となる。また、光学データと共に、その光学データを用いて多重歪み補正を行う動作プログラムも記録し、その動作プログラムに従って多重歪み補正を行うようにしても良い。
【０１９５】
［付記］
１．請求項３において、幾何学的多重歪み補正は、まず第２の歪み補正を行った後に第１歪み補正を行う。
【０１９６】
２．請求項５において、幾何学的多重歪み補正は、まず第２歪み後段補正を行い、次に第２歪み前段補正を行い、最後に第１歪み補正を行う。
【０１９７】
３．異なる視点からの像を得るための複数の対物光学系と、前記複数の対物光学系の像を伝送する像伝送光学系とによる複数の像が結像させる撮像素子からの出力信号に対して、前記複数の対物光学系による第１の幾何学的歪みと、前記像伝送光学系による第２の幾何学的歪みが重なった歪みである多重歪みを補正する幾何学的多重歪み補正手段を設けた画像処理装置。
【０１９８】
４．異なる視点からの像を得るための複数の対物光学系と、前記複数の対物光学系の像を伝送する像伝送光学系とによる複数の像が結像させる撮像素子からの出力信号に対して、前記複数の対物光学系による第１の幾何学的歪みと、前記像伝送光学系による第２の幾何学的歪みが重なった歪みである多重歪みを補正するデータを記録した記録媒体。
【０１９９】
【発明の効果】
以上説明したように本発明によれば、内視鏡の先端部に１対の対物光学系と、該対物光学系による光学像を伝送する像伝送光学系とからなる撮像光学系を設けて撮像素子に結像するようにしているので、挿入部の外径を細くできると共に、１対の対物光学系により発生する第１の歪みと像伝送光学系により発生する第２の歪みとが重なった多重歪みを補正する多重歪み補正手段を設けているので、精度の良い計測ができると共に、多重歪みを補正することにより立体視した場合の疲労感を軽減することができる。
【図面の簡単な説明】
【図１】本発明の第１の実施の形態を示す計測内視鏡装置の全体を示す斜視図。
【図２】図１の計測内視鏡装置の構成を示すブロック図。
【図３】内視鏡の挿入部の先端部を拡大して示す斜視図。
【図４】図３のＡ−Ａ断面図。
【図５】幾何学的歪みの補正前の画像と補正後の画像のピクセル配置図。
【図６】補正前の画像ピクセルと補正後の画像ピクセル図。
【図７】多重歪み補正を含む座標変換の処理の内容を示すフローチャート図。
【図８】ステレオ計測処理の内容を示すフローチャート図。
【図９】図８のステレオ計測処理でコールされるサブルーチンの内容を示すフローチャート図。
【図１０】図８の計測処理でコールされるサブルーチンの内容を示すフローチャート図。
【図１１】計測内視鏡装置を用いて深さ計測状態のパイプを示す斜視図。
【図１２】図１１のパイプの計測部分の拡大図。
【図１３】計測内視鏡装置を用いて長さ計測状態のパイプを示す斜視図。
【図１４】計測内視鏡装置を用いてくぼみ計測状態のタービンブレードを示す斜視図。
【図１５】図１０の計測処理でコールされるサブルーチンを示し、図１５（Ａ）は、３次元座標解析処理のフローチャート図、図１５（Ｂ）は、パターンマッチング処理のフローチャート図。
【図１６】計測内視鏡装置における３次元座標解析処理の基本原理を説明するためのｘ，ｙ，ｚ軸をもつ３次元座標系上の右，左２画像位置関係を示す図。
【図１７】計測内視鏡装置による左右の２画面の画像をＬＣＤ上に表示したステレオ計測画面を示す図。
【図１８】本発明の第２の実施の形態における挿入部の先端部を示す斜視図。
【図１９】図１８のＢ−Ｂ断面図。
【図２０】光学アダプタを生産測定治具のマスタ撮像ユニットに接続して光学データを採集する様子を示す斜視図。
【図２１】ステレオ計測処理の内容を示すフローチャート図。
【図２２】図２１のステレオ計測処理でコールされるサブルーチンの内容を示すフローチャート図。
【図２３】光学アダプタをマスタ撮像ユニットに接続して撮像された白色チャートの画像を示す図。
【図２４】撮像された白色チャートの画像を表示した図。
【図２５】計測内視鏡装置およびマスタに対する視野形状パターンを与える輝度波形であって、図２５（Ａ）は、画像の垂直方向のＰＶ，ＰＶ’の輝度変化、図２５（Ｂ）は、画像の水平方向のＰＨ，ＰＨ’の輝度変化を示す図。
【図２６】本発明の第３の実施の形態における挿入部の先端部の構造を示す断面図。
【図２７】本発明の第４の実施の形態における座標変換の処理内容を示すフローチャート図。
【符号の説明】
１…計測内視鏡装置
２…内視鏡
３…コントロールユニット
４…リモコン
５…ＬＣＤ
６…ＦＭＤ
９…ＣＣＵ
１８…計測処理部
１８ａ…ＣＰＵ
２０…挿入部
２１…先端部
２６Ｒ，２６Ｌ…対物光学系
２７…像伝送光学系
３１…パソコン
３３…メモリカード[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an endoscope apparatus that forms images of an object to be measured by a plurality of objective optical systems at different positions on an image sensor provided in an endoscope, and performs measurement using the images.
[0002]
[Prior art]
As a measurement endoscope apparatus, for example, there is a first conventional example disclosed in Japanese Patent Application Laid-Open No. Hei 10-248806.
[0003]
This conventional example is an example in which respective images of a plurality of objective optical systems provided at the distal end of an insertion section are directly formed on an image sensor, and optical distortion correction of images is performed for each optical system.
[0004]
When the size of the imaging element is reduced in order to reduce the outer diameter of the insertion portion, the interval between the objective optical systems is reduced, the measurement accuracy is reduced, and the obtained three-dimensional effect is reduced.
[0005]
JP-A-11-109257 discloses a second conventional example which solves the problem of the conventional example. In the second conventional example, in order to solve the above-described problem, an image transmission optical system that collectively transmits images of a plurality of objective optical systems is provided between the objective optical system and the imaging device. As a result, a small-sized image sensor can be adopted without reducing the distance between the objective optical systems.
[0006]
[Problems to be solved by the invention]
Since an image transmission optical system that transmits a plurality of images collectively is interposed, distortion due to the image transmission optical system itself is superimposed in addition to optical distortion due to each objective optical system.
[0007]
In the distortion correction based on the optical center on the image of each objective optical system, sufficient distortion correction cannot be performed, and there is a disadvantage that measurement accuracy is reduced and fatigue in stereoscopic viewing (stereo observation) is increased.
[0008]
(Object of the invention)
The present invention has been made in view of the above points, and an object of the present invention is to provide an endoscope apparatus capable of reducing the outer diameter of an insertion portion, having high measurement accuracy, and reducing fatigue when performing stereoscopic viewing. I do.
[0009]
[Means for Solving the Problems]
A plurality of objective optical systems for obtaining images from different viewpoints,
An image transmission optical system for transmitting images of the plurality of objective optical systems,
An imaging element that forms a plurality of images transmitted to the image transmission optical system,
A processing unit that receives a signal from the image sensor and generates a video signal;
In an endoscope device comprising:
In order to correct the optical distortion of the video signal, the first geometric distortion by the plurality of objective optical systems and the multiple distortion, which is a distortion obtained by overlapping the second geometric distortion by the image transmission optical system, are used. By providing a geometric multiple distortion correction means for correcting, the geometric multiple distortion correction taking into account both the geometric distortion due to the image transmission optical system and the geometric distortion due to the plurality of objective optical systems is performed, and inserted. The outer diameter of the part can be reduced, the measurement accuracy is high, and fatigue when performing stereoscopic vision can be reduced.
[0010]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0011]
(First Embodiment)
FIGS. 1 to 17 relate to the first embodiment of the present invention, FIG. 1 is a perspective view showing the entire configuration of the measurement endoscope apparatus according to the first embodiment of the present invention, and FIG. 3 shows the appearance near the distal end of the insertion portion, FIG. 4 shows the structure taken along the line AA in FIG. 3, and FIG. 5 shows the image before correction of the geometric distortion and the correction. FIG. 6 shows an image pixel before correction and an image pixel after correction, FIG. 7 shows the contents of coordinate conversion processing including multiple distortion correction, and FIG. 8 shows stereo measurement. 9 shows the contents of a subroutine called in the stereo measurement processing of FIG. 8, FIG. 10 shows the contents of a subroutine called in the measurement processing of FIG. 8, and FIG. 11 shows the measurement endoscope apparatus. FIG. 12 is a perspective view of the pipe in a depth measurement state using FIG. FIG. 13 is an enlarged view of a measurement portion, FIG. 13 is a perspective view of a pipe in a length measurement state using a measurement endoscope device, and FIG. 14 is a perspective view of a turbine blade in a dent measurement state using the measurement endoscope device. FIG. 15 is a flowchart of a subroutine called in the measurement processing of FIG. 10. FIG. 15 (A) is a flowchart of three-dimensional coordinate analysis processing, and FIG. 15 (B) is a flowchart of pattern matching processing. FIG. 16 shows a right and left image positional relationship on a three-dimensional coordinate system having x, y, and z axes for explaining a basic principle of a three-dimensional coordinate analysis process in the measurement endoscope apparatus. 4 shows a stereo measurement screen in which two left and right images by the measurement endoscope apparatus are displayed on an LCD.
[0012]
As shown in FIG. 1, an endoscope apparatus 1 for measurement according to a first embodiment of the present invention includes an endoscope 2 having an elongated and flexible insertion portion 20, and an insertion portion of the endoscope 2. The control unit 3 is a control device having a storage unit for storing the control unit 20, a remote controller 4 for performing operations necessary for performing various operation controls of the entire device, an endoscope image and operation control contents (for example, processing A liquid crystal monitor (hereinafter, referred to as LCD) 5 which is a display device for displaying menus, etc., and a face which enables stereoscopic viewing as a normal endoscope image or a pseudo stereo image using the endoscope image. It is mainly composed of a mount display (hereinafter referred to as FMD) 6, an FMD adapter 6a for supplying image data to the FMD 6, and the like.
[0013]
The insertion portion 20 is configured by connecting a rigid distal end portion 21 in order from the distal end side, for example, a bending portion 22 that can be bent vertically and horizontally, and a flexible tube portion 23 having flexibility. As shown in FIG. 3, two illumination windows 24 and observation windows 25 are provided.
[0014]
As shown in FIG. 4, a pair of objective optical systems that enable stereoscopic viewing, that is, the right image objective lens 26R and the left image objective lens 26L are separated from each other so that the observation window 25 has parallax. It is attached. That is, images are obtained from different viewpoints. An optical image of an object to be inspected, such as an inspection of a pipe inner surface or a turbine blade, is inspected by a pair of objective lenses 26R and 26L through a common image transmission optical system 27 and a solid-state image sensor such as a CCD (simply referred to as an image sensor). 28.
[0015]
In the present embodiment, as described in the second conventional example, image transmission optics that collectively transmits images by a pair of objective optical systems between a pair of objective optical systems and the imaging device 28. By interposing the system 27, a small-sized image pickup device 28 is employed without reducing the distance between the pair of objective optical systems, so that highly accurate three-dimensional (stereo) measurement can be performed.
[0016]
As shown in FIG. 2, the control unit 3 includes an endoscope unit 8, a camera control unit (hereinafter referred to as CCU) 9 for performing processing for generating a video signal, and a control unit 10 for performing measurement control and the like. The base end of the insertion section 20 is connected to the endoscope unit 8.
[0017]
The endoscope unit 8 includes a light source device (not shown) for supplying illumination light necessary for observation, and a bending device (not shown) for bending the bending portion 22 constituting the insertion section 20. .
[0018]
The CCU 9 receives an image signal output from an image sensor 28 built in the distal end portion 21 of the insertion section 20. This image pickup signal is subjected to signal processing in the CCU 9, converted into a video signal such as an NTSC signal, and supplied to the control unit 10.
[0019]
The control unit 10 includes an audio signal processing circuit 11, a video signal processing circuit 12 to which the video signal is input, a ROM 13, a RAM 14, a PC card interface (hereinafter, referred to as a PC card I / F) 15, and a USB interface (hereinafter, a USB interface). , USB @ I / F) 16 and an RS-232C interface (hereinafter, referred to as "RS-232C @ I / F") 17 and the like, and various functions are executed based on a main program to perform operation control and measurement processing. And a CPU 18a that forms a measurement processing unit 18 that performs the measurement are connected to each other via a bus.
[0020]
The measurement processing unit 18 includes a CPU 18a that performs a measurement process, a ROM 18b that stores an operation program and the like of the CPU 18a, and a RAM 18c that is used as a work area of the CPU 18a or used as a memory for storing necessary data. And these are connected to the bus.
[0021]
Further, as described later, when performing measurement (for example, by reading optical data recorded on the memory card 33), the CPU 18a performs geometric multiplexing of the imaging optical system, that is, the objective lenses 26R and 26L and the image transmission optical system 27. Coordinate transformation including distortion correction is performed so that highly accurate measurement can be performed. The function of the multiple distortion correction processing 18d indicated by a two-dot chain line in the CPU 18a will be described later.
[0022]
The RS-232C @ I / F 17 is connected to the CCU 9, the endoscope unit 8, and a remote controller (hereinafter, abbreviated as a remote controller) 4 for controlling and operating the CCU 9, the endoscope unit 8, and the like. . Accordingly, communication necessary for controlling the operation of the CCU 9 and the endoscope unit 8 based on the operation of the remote controller 4 is performed.
[0023]
The USB I / F 16 is an interface for electrically connecting the control unit 3 to the personal computer 31. By connecting the control unit 3 to the personal computer 31 via the USB I / F 16, the personal computer 31 can control various instructions such as an instruction to display an endoscope image and image processing at the time of measurement. And control information and data necessary for various processes between the control unit 3 and the personal computer 31 can be input and output.
[0024]
A so-called memory card, which is a recording medium such as a PCMCIA memory card 32 or a compact flash (R) memory card 33, can be freely inserted into and removed from the PC card I / F 15.
[0025]
When the memory cards 32 and 33 are mounted on the PC card I / F 15, the control of the CPU 18 allows the control cards to fetch or control data such as control processing information and image information stored in the memory cards 32 and 33. Data such as processing information and image information can also be recorded on the memory cards 32 and 33.
[0026]
Further, in the present embodiment, a personal computer (abbreviated as a personal computer) 31 is connected to the measurement endoscope apparatus 1, and distortion correction is performed by the objective lenses 26 R and 26 L provided at the distal end of the endoscope 2 and the image transmission optical system 27. Is performed, the optical data is recorded in the memory card 33, and the measurement processing unit 18 reads the optical data when performing measurement such as three-dimensional measurement.
[0027]
Then, by referring to the optical data, the CPU 18a of the measurement processing unit 18 performs the coordinate conversion in which the distortion is corrected by the objective lenses 26R and 26L and the image transmission optical system 27, that is, the imaging optical system, and the image is displayed on the LCD 5. And so on.
[0028]
In the video signal processing circuit 12, an operation menu for generating a video signal from the CCU 9 under the control of the CPU 18a so as to display a synthesized image obtained by synthesizing the endoscopic image supplied from the CCU 9 and an operation menu by graphics. Based on the display signal, a synthesizing process with a display signal, a process necessary for displaying the image on the screen of the LCD 5, and the like are supplied to the LCD 5.
[0029]
In addition, the video signal processing circuit 12 can also perform processing for simply displaying an image such as an endoscope image or an operation menu alone. Therefore, an endoscope image, an operation menu image, a composite image of the endoscope image and the operation menu image, and the like are displayed on the screen of the LCD 5.
[0030]
When the image correction instruction operation is performed, the video signal output from the CCU 9 to the video signal processing circuit 12 is converted into a digital signal by an A / D converter (not shown) inside the video signal processing circuit 12. Then, it is taken into the CPU 18a constituting the measurement processing unit 18 and subjected to multiple distortion correction. Then, the D / A converter in the video signal processing circuit 12 returns to an analog video signal again and is output to the LCD 5. Then, the image with the distortion corrected is displayed.
[0031]
The audio signal processing circuit 11 includes an audio signal generated by collecting sound by the microphone 34 and recorded on a recording medium such as a memory card, an audio signal obtained by reproducing a recording medium such as a memory card, or the CPU 18a. The generated audio signal is supplied. The audio signal processing circuit 11 performs processing such as amplification necessary for reproducing the supplied audio signal and outputs the processed audio signal to the speaker 35. As a result, sound is output from the speaker 35.
[0032]
Further, the CPU 18a controls the various circuit sections and the like so as to execute processing corresponding to functions other than the above-described distortion correction processing by executing a program stored in the ROM 18b, and also controls the operation of the entire system. Do.
[0033]
As shown in FIG. 1, on one surface of the remote controller 4, a joystick 4a for instructing a bending operation of the bending section 22, a lever switch 4b for performing various menu operations displayed graphically and a pointer moving operation for performing measurement, A freeze switch 4c for instructing the LCD 5 to display a still image, a store switch 4d for recording a still image on the memory card 33 and the like, a measurement execution switch 4e used when executing measurement software, and a connector unit 4f to which the FMD6 adapter 6a is connected. Etc. are provided.
[0034]
FIG. 3 is an enlarged perspective view of the distal end portion 21 of the insertion section 20 of the endoscope 2, and FIG. 4 is a sectional view taken along line AA of FIG.
[0035]
As shown in FIG. 3, two illumination windows 24 and an observation window 25 that enables stereoscopic viewing are provided on the distal end surface of the distal end portion 21 of the endoscope 2.
[0036]
As shown in FIG. 4, the observation window 25 provided in the distal end portion main body 29 constituting the distal end portion 21 is closed by a cover glass 30, and a pair of objective optical systems, that is, a right An image objective optical system 26R and a left image objective optical system 26L are attached.
[0037]
The lens frame 36 extends rearward, and a front side (optical system) 27a of the common image transmission optical system 27 is attached. The image sensor 28 is fixed to the image sensor fixing frame 37 housed and fixed in the hole on the rear end side of the lens frame 36. A rear stage (optical system) 27b of the image transmission optical system 27 is mounted on the front side of the imaging surface of the imaging device 28 via a lens frame.
[0038]
Further, the outer periphery of the front end side main body 29 on the front end side is covered with a cylindrical cover member 38, and the cover member 38 is fixed to the front end main body 29 with screws. In addition, an O-ring for sealing is interposed between the cover member 38 and the distal end main body 29 to have a watertight structure.
[0039]
The images formed by the objective optical system 26R and the objective optical system 26L are formed on the image sensor 28 via the image transmission optical system 27 at different positions on the left and right. That is, a right imaging optical system, which is an optical system including the objective optical system 26R and the image transmission optical system 27, and a left imaging optical system, which is an optical system including the objective optical system 26L and the image transmission optical system 27, are configured.
[0040]
Next, the operation of the present embodiment will be described.
[0041]
In the endoscope 2 applied to the present measurement endoscope apparatus 1, as shown in the following (a1) to (d), optical data of an imaging optical system unique to each endoscope 2 is measured. The optical data is recorded on a recording medium, for example, a memory card 33. This specific optical data corresponds to the imaging optical system on a one-to-one basis, and is handled as one combination after shipping.
[0042]
The above specific optical data is
(A1) Geometric distortion correction table for two objective optical systems
(A2) Geometric distortion correction table of image transmission optical system
(B) The focal length of each of the left and right imaging optical systems
(C) Distance between principal points of left and right imaging optical systems
(D) Optical axis position coordinates on each image of the left and right imaging optical systems
It is.
[0043]
The personal computer 31 is connected to the measurement endoscope apparatus 1 after the above-mentioned specific optical data has been collected, and the following processes (1) to (5) can be performed to measure various dimensions. That is,
(1) The optical data (a1) to (d) is read from the memory card 33. (2) The endoscope 2 captures an image of an object to be measured, and captures the image.
[0044]
(3) The captured image is coordinate-transformed based on the optical data (a1) to (d).
[0045]
(4) Three-dimensional coordinates of an arbitrary point are obtained by matching imaging data based on the image after the coordinate conversion.
[0046]
(5) Various dimensions are measured based on the three-dimensional coordinates.
[0047]
When the measurement endoscope apparatus 1 captures optical data during production, a personal computer 31 is connected to the measurement endoscope apparatus 1 as shown in FIG. 2 and a chart not shown here (second embodiment) (See reference numeral 54 in FIG. 20).
[0048]
In the present measurement endoscope apparatus 1, the above-described processing can be performed to perform measurement with high accuracy on the object to be measured.
[0049]
Next, each optical data specific to the above-described imaging optical system will be specifically described.
[0050]
(A1), (a2) Regarding the geometric distortion correction table,
Generally, an image obtained by a lens system has optical distortion. When performing measurement, this distortion causes a large error. Therefore, this distortion can be removed by performing coordinate transformation. The coordinate conversion may be performed with the center of the optical axis as the center. For more accurate correction, the center of the geometric distortion of the optical system may be used.
[0051]
Also, the geometric distortion correction tables for the two images may be provided separately for the right image and the left image, or the two may be combined into one table. Hereinafter, the correction table in the case of one table will be described with reference to FIGS.
[0052]
5 and 6, points p1 to p4 on the imaging screen 40 indicate pixels before coordinate conversion. When coordinates of p1 to p4 are transformed by f (x, y), they are p1 'to p4'. The coordinates giving p1 'to p4' at this time are not necessarily integers but are obtained as real-valued coordinates. To display on the converted screen 41 on the liquid crystal monitor after the conversion, the coordinates (X, Y) of the converted pixel P (X, Y) must be converted into an integer value in pixel units.
[0053]
The correction for converting the coordinates into an integer value is performed by weight tables W1 to W4. That is, for each pixel of the conversion screen, the image data of four pixels whose coordinates correspond to the optical geometric distortion on the imaging screen is multiplied by the ratio of the weight tables W1 to W4, and the pixel of the conversion screen pixel is obtained. Data P (X, Y) is obtained.
[0054]
Therefore, an expression of coordinate conversion including the processing of the multiple distortion correction (specifically, the second distortion correction, the first distortion right correction, the first distortion left correction) is as follows.
[0055]
Second distortion correction:
f_2x(U, v) ≡k_x(A₀₀+ A₁₂uv²+ A₁₄uv⁴+ A₁₆uv⁶+ A₃₀u³+ A₃₂u³v²
+ A₃₄u³v⁴+ A₅₀u⁵+ A₅₂u⁵v²+ A₇₀u⁷) + C_x(1-1)
f_2y(U, v) ≡k_y((B₀₀+ B₂₁u²v + b₄₁u⁴v + b₆₁u⁶v + b₀₃v³+ B₂₃u²v³
+ B₄₃u⁴v³+ B₀₅v⁵+ B₂₅u²v⁵+ B₀₇u⁷) + C_y(1-2)
x₂= F_2x(Xc_x, Y-c_y) (1-3)
y₂= F_2y(Xc_x, Y-c_y) (1-4)
First distortion right correction:
f_Rx(U, v) ≡k_Rx(A_R00+ A_R12uv²+ A_R14uv⁴+ A_R16uv⁶+ A_R30u³+ A_R32u³v²
+ A_R34u³v⁴+ A_R50u⁵+ A_R52u⁵v²+ A_R70u⁷) + C_Rx(1-5)
f_Ry(U, v) ≡k_Ry(B_R00+ B_R21u²v + b_R41u⁴v + b_R61u⁶v + b_R03v³+ B_R23u²v³
+ B_R43u⁴v³+ B_R05v⁵+ B_R25u²v⁵+ B_R07u⁷) + C_Ry(1-6)
x_R’= F_Rx(X₂-C_Rx, Y₂-C_Ry) (1-7)
y_R’= F_Ry(X₂-C_Rx, Y₂-C_Ry) (1-8)
First distortion left correction:
f_Lx(U, v) ≡k_Lx(A_L00+ A_L12uv²+ A_L14uv⁴+ A_L16uv⁶+ A_L30u³+ A_L32u³v²
+ A_L34u³v⁴+ A_L50u⁵+ A_L52u⁵v²+ A_L70u⁷) + C_Lx(1-9)
f_Ly(U, v) ≡k_Ly(B_L00+ B_L21u²v + b_L41u⁴v + b_L61u⁶v + b_L03v³+ B_L23u²v³
+ B_L43u⁴v³+ B_L05v⁵+ B_L25u²v⁵+ B_L07u⁷) + C_Ly(1-10)
x_L’= F_Lx(X₂-C_Lx, Y₂-C_Ly) (1-11)
y_L’= F_Ly(X₂-C_Lx, Y₂-C_Ly) (1-12)
The signs in the above equation have the following meanings.
[0056]
x, y: coordinates before correction
x_R’, Y_R’: Right screen coordinate after correction
x_L’, Y_L’: Left screen coordinate after correction
c_x, C_y: Second distortion center coordinates
c_Rx, C_Ry: 1st distortion right center coordinate
c_Lx, C_Ly: 1st distortion left center coordinate
k_x, K_y, A_ij, B_ij: Second distortion correction coefficient
k_Rx, K_Ry, A_Rij, B_Rij: 1st distortion right correction coefficient
k_Lx, K_Ly, A_Lij, B_Lij: 1st distortion left correction coefficient
Note that k_xCoefficient a excluding k such as_ij, B_ij, A_Rij, B_Rij, A_Lij, B_Lij, C_x, C_y, C_Rx, C_Ry, C_Lx, C_LyIs determined from the linearity of the grid image. Also, k_x, K_yAnd the like are coefficients for adjusting the magnification of the two images and are functions of the focal lengths fR and fL.
[0057]
Using the above equation, the x, y coordinates of p1 to p4 are substituted to obtain coordinates (x ', y') that give p1 '(x', y ') to p4' (x ', y'). The values of x 'and y' are not necessarily integers as described above, but are corrected by the weight tables W1 to W4 to obtain pixel data for the integer converted coordinates (X, Y).
[0058]
Therefore, the coordinates of the four points p1 'to p4' on the conversion screen for providing the converted pixel data P (X, Y) are defined as (x ', y'), and the four points p1 'to p4 on the conversion screen. The x coordinate of the coordinate (x, y) of the upper left point p1 among the coordinates (x, y) to (x + 1, y + 1) of the four points on the imaging screen (original image) corresponding to '′ is QX (X, Y) And the y coordinate is QY (X, Y), which is first recorded in the memory card 33 as a coordinate reference table of the geometric distortion correction table.
[0059]
5 and 6, the cases of the corrected right image coordinates xR 'and yR' and the left image coordinates xL 'and yL' are collectively indicated by x 'and y'.
[0060]
The converted pixel data P (X, Y) of the coordinates (X, Y) given by the converted pixel unit integer value is obtained from p1 'to p4' and W1 to W4. However, as shown in FIG. 6, assuming that dn indicates the distance from p1 'to p4' to P (X, Y),
S = d2 · d3 · d4 + d1 · d3 · d4 + d1 · d2 · d4
+ D1, d2, d3 (2)
further,
W1 = d2 · d3 · d4 / S
W2 = d1 · d3 · d4 / S
W3 = d1 · d2 · d4 / S
W4 = d1 · d2 · d3 / S (3)
And And the value of P (X, Y) is
P (X, Y) = W1 × p1 ′ + W2 × p2 ′ + W3 × p3 ′ + W4 × p4 ′ (4)
Required by
[0061]
W1, W2, W3, and W4 are weight tables for each pixel point (X, Y) on each conversion screen together with the coordinate reference tables QX (X, Y) and QY (X, Y). Recorded in.
[0062]
Next, regarding (b) to (d),
(B) Find the focal length fR of the right imaging optical system and the focal length fL of the left imaging optical system. (C) The distance L between the principal points of the right imaging optical system and the left imaging optical system is obtained and recorded.
[0063]
(D) The optical axis position coordinates XR and YR on the image of the right imaging optical system and the optical axis position coordinates XL and YL on the image of the left imaging optical system are obtained.
[0064]
The coordinate conversion process including the above-described processes of the second distortion correction, the first distortion right correction, and the first distortion left correction is performed as shown in FIG.
[0065]
When the coordinate conversion process starts, the CPU 18a fetches the second distortion correction data from the second distortion correction table as shown in step S1, and the CPU 18a executes the above equations (1-1) to (1--1) as shown in step S2. The second distortion correction process is performed according to 4).
[0066]
In the next step S3, the CPU 18a fetches the data of the first distortion correction from the first distortion correction table, and performs the processing of the first distortion correction as shown in step S4.
In this case, the first distortion correction processing is performed by the above equations (1-5) to (1-8) and (1-9) to (1-12), and the coordinate conversion ends.
[0067]
Thereafter, the coordinates (X, Y) of the integer value are calculated using the weight tables W1 to W4 as described above.
[0068]
Optical data such as a table including the distortion correction and focal lengths of the objective optical system and the image transmission optical system 27 are recorded on the memory card 33. When measurement is performed by the measurement endoscope apparatus 1 as described below, Read and used.
[0069]
The stereo measurement processing as the measurement operation by the measurement endoscope apparatus 1 according to the present embodiment configured as described above will be described with reference to the flowcharts of FIGS. 8, 9, 10, and 15, and FIGS. explain.
[0070]
When the stereo measurement process starts, the corrected image information data set in the first step S11 of FIG. 8 is read as original data, and the process proceeds to step S12.
[0071]
In step S12, image data is fetched. That is, an object to be measured is imaged and taken into a memory as original image data. Then, in step S13, an image correction process of a subroutine described later (see FIG. 9) is executed to perform image correction of the original image. That is, the original image data captured in step S12 is subjected to coordinate transformation based on the original table, and measurement image data for performing measurement is obtained.
[0072]
Subsequently, in step S14, based on the measurement image data, a subroutine measurement process (see FIG. 10), which will be described later, is executed, and three-dimensional coordinates are calculated and the selected measurement is performed. finish.
[0073]
The image correction processing of the subroutine called in step S13 (see FIG. 8) will be described with reference to the flowchart of FIG.
[0074]
The pixel data P (X, Y) on the measurement image can be obtained by the processing of this subroutine. First, in step S21, original data, which is a correction table created at a factory, is fetched. In step S22, the coordinates of the corrected image, that is, the coordinates of the measured image are set as X and Y.
[0075]
In step S23, the coordinates of the original image necessary to obtain the corrected coordinates X and Y are set as reference coordinates, and the corrected coordinates X and Y are calculated using QX (X, Y) and QY (X, Y) created in the flowchart of FIG. Get. Then, in step S24, the weight table is fetched, and in step S25, pixel data P (X, Y) on the coordinates after correction is obtained by the following equation.
[0076]
That is,

Required by
[0077]
Subsequently, in step S26, a corrected image range is set, and this routine ends. Note that the image correction processing of this subroutine is performed on the left and right images, and the original image is converted into a measurement image.
[0078]
Next, the measurement processing of the subroutine called in step S14 of FIG. 8 will be described with reference to the flowchart of FIG.
[0079]
First, in step S171, three measurement methods are selected on a menu displayed on the LCD 5. One of them is depth measurement. When the distance between a point and a plane is measured, for example, as shown in the perspective view of FIG. (See Reference).
[0080]
When this measurement method is selected, the process jumps to step S172. The other one is a length measurement. When measuring a distance between two points, for example, as shown in a perspective view of FIG. 13, three-dimensional coordinates of two points of a pipe 42 which is a three-dimensional object to be measured are used. This is applied when obtaining the distance d between two points.
[0081]
When this measurement method is selected, the process jumps to step S190. Another one is dent measurement, and when measuring the distance between a point and a straight line, for example, as shown in the perspective view of FIG. It is applied when measuring etc. When this measurement method is selected, the process jumps to step S200.
[0082]
When jumping to step S172, in steps S173 to S175, three reference points PT for setting a reference plane are set. For example, in the case of the pipe 51 of the measured object shown in FIG. 11, measurement reference points PT1, PT2, PT3 are set. Then, the three-dimensional coordinates of the above three points are obtained by a subroutine three-dimensional coordinate analysis process described later (see FIG. 15A).
[0083]
In step S176, the point PT instruction value is incremented, and in step S177, the designated value is set as a reference plane of the depth of the plane formed by the three points. In the case of FIG. 11, a plane M1 including the points PT1, PT2, and PT3 is set. Further, a measurement point, for example, PT4 is set in step S178, and a subroutine three-dimensional coordinate analysis process (see FIG. 15A) is called and executed in S179 to obtain the coordinates of the measurement point. In the case of FIG. 12, the coordinates of the point PT4 are obtained.
[0084]
In step S180, the distance from the reference plane to the measurement point, in the case of FIG. 12, the distance h that gives the corrosion depth from the measurement point PT4 to the reference plane M1 is calculated, and this routine ends.
[0085]
Next, when jumping to step S190, a first measurement point is set in step S191, and a subroutine three-dimensional coordinate analysis process is executed in step S192 to obtain the coordinates of the first measurement point. Further, a second measurement point is set in step S193, and a subroutine three-dimensional coordinate analysis process is executed in step S194 to obtain the coordinates of the second measurement point. Thereafter, in step S195, the distance d between the two measurement points is calculated from the coordinates of the first and second measurement points, and the present subroutine ends.
[0086]
In the above processing, when the object to be measured is the pipe 42 shown in FIG. 13 and the first and second measurement points are PT5 and PT6, the distance d between the two measurement points is obtained by the above measurement. Can be
[0087]
Further, when jumping to step S200, a first reference point is set at one end of the depression in step S201, and a subroutine three-dimensional coordinate analysis process is executed in step S202 to obtain the coordinates of the first reference point. Further, a second reference point is set at the other end of the depression in step S203, and a subroutine three-dimensional coordinate analysis process is executed in step S204 to obtain the coordinates of the second reference point. In step S205, a reference straight line connecting the two points is set from the coordinates of the first and second reference points.
[0088]
In step S206, a measurement point of the depression is set. In step S207, a subroutine three-dimensional coordinate analysis process is executed to determine the coordinates of the measurement point. In step S208, the length h of the perpendicular line between the straight line connecting the reference points and the measurement point is calculated as the depression amount, and the routine ends.
[0089]
When the object to be measured is the turbine blade 43 shown in FIG. 14 in the processing of the above subroutine, the distance h between the reference straight line connecting the first and second reference points PT7 and PT8 and the measurement point PT9 of the depression is obtained. Will be.
[0090]
Next, the three-dimensional coordinate analysis processing of the subroutine called in steps S175, S192, S202 and other steps will be described with reference to the flowchart of FIG.
[0091]
In step S211, a pattern matching process of a subroutine is executed to detect a matching point that is a corresponding point of the two left and right images (stereo images). In step S212, the amount of displacement between the left and right images is obtained from the coordinates of the corresponding point. In step S213, the three-dimensional coordinates of the target point are calculated, and this routine ends. Here, the basic principle of the three-dimensional coordinate analysis processing in the above subroutine will be described with reference to FIG.
[0092]
FIG. 16 is a diagram showing a right-left two-image positional relationship on a three-dimensional coordinate system having x, y, z axes, in which a point P of a subject is a right image plane 28R and a left image plane 28L of the image sensor 28. The upper image is shown. In FIG. 16, points OR and OL are the principal points of the optical system, distance f is the focal length, points QR and QL are the imaging positions of point P, and distance L is the distance between point OR and point OL. And
[0093]
In FIG. 16, the following equation is established from the straight line QR-OR. That is,
x / x_R'= {Y- (L / 2)} / ｛y_R'-(L / 2)｝ = z / (-f) (6)
Further, the following equation is established from the straight line QL-OL. That is,
x / x_L′ = {Y + (L / 2)} / ｛y_L'+ (L / 2)｝ = z / (-f) (7)
By solving this equation for x, y, z, the three-dimensional coordinates of the point P can be obtained.
[0094]
Note that, actually, due to the effect of the image transmission optical system 27, the light beams of the two left and right images are bent and the distance between the right imaging surface 28R and the left imaging surface 28L becomes smaller, but here, in order to simplify the figure, The illustration does not show the effect of the image transmission optical system 27.
[0095]
Next, the pattern matching processing of a subroutine called in step S211 of the three-dimensional coordinate analysis processing of FIG. 15A will be described with reference to the flowchart of FIG.
[0096]
In this routine, pattern matching for detecting a corresponding point between two images is performed. FIG. 17 shows a stereo measurement screen 64 in which images on the left and right two screens 65 and 66 are displayed on the LCD 5, and shows that the image point PP1 on the left screen 65 corresponds to the image point PP2 on the right screen 66. ing.
[0097]
First, in steps S221 and S222, a pattern area for setting the size of a pattern to be subjected to pattern matching is narrowed down. In the example of the present embodiment, a pattern area corresponding to the value k is set. That is,
When k = 1, the pattern area is 35 × 35 (pixel),
When k = 2, the pattern area is 23 × 23 (pixels),
When k = 3, the pattern area is 11 × 11 (pixels),
The value k is switched from small to large, and the area is narrowed from large to small, so that the accuracy of corresponding point detection is increased.
[0098]
In step S223, a search range is set. That is, the area of the right image to search for the pattern is determined. The search range is set in the case where the epipolar line is set within ± 5 pixels in consideration of an error, in the case where the epipolar line is set within ± 7 pixels horizontally on the monitor screen, or in the case where the outline is manually designated on the screen. In some cases, it is within ± 10 pixels around the matching point. Note that the above ± 10 pixels is an optimal value in consideration of a manual error.
[0099]
In steps S224 to S226, pattern matching in the set search range is performed. In this pattern matching, corresponding points are detected by normalized cross-correlation, and coordinates (X, Y) having the largest normalized cross-correlation coefficient are set as the corresponding points.
[0100]
In step S227, the value k is incremented, patterns are narrowed down in accordance with the value k, and corresponding points are detected.
[0101]
In step S228, a matching point is set. At this time, the value of the normalized cross-correlation may be displayed on a monitor screen, and this may be used as a measure of the reliability of the matching. If the value of the normalized cross-correlation (−1 to 1) is smaller than a predetermined value, the process may be switched to a manual matching process.
[0102]
The following equation is generally used as the normalized cross-correlation function M (u, v) used for the pattern matching. That is, t (x, y) is used as a template, g (x, y) is used as image data, t 'is used as the average luminance of the template, and g' is used as the average luminance of the image.
M (u, v) = ｛ΣΣ_s(G (x + u, y + v) -g ') (t (x, y) -t')}
/ ｛ΣΣ_s(G (x + u, y + v) -g ')²× ΣΣ_s(T (x, y) -t ')²｝^1/2(8)
Is applied. Here ΣΣ_sRepresents the sum of pixels.
[0103]
As an example of measurement according to the present embodiment, for example, measurement is performed on a corrected image. In this case, when the measurement execution switch 4 e is operated by the remote controller 4, the corrected left and right images are displayed on the LCD 5.
[0104]
Further, by operating the lever switch 4b of the remote controller 4 to input an instruction of a point to be measured on each of the left and right images, stereo measurement can be performed.
[0105]
It is also possible to create a corrected image and perform stereoscopic viewing. In this case, when the measurement execution switch 4e of the remote controller 4 is operated, the corrected left and right images are displayed on the LCD 5, and the corrected left and right image signal is input to the FMD adapter 6a. After a process of separating the video signal into left and right video signals, the video signal is input to the FMD 6.
[0106]
By wearing the FMD 6 and stereoscopically viewing, the user can observe the left and right images of good image quality in which distortion has been corrected on the left and right, respectively, which places a greater burden on the user than observing the uncorrected left and right images. Without this, stereoscopic vision can be achieved, and user fatigue can be greatly reduced.
[0107]
Although the processing of the image data in the above-described embodiment has been described as a case where a monochrome image is used as an image for the sake of simplicity, the processing may be extended to a color image using R, G, and B signals. Of course, it is equally effective.
[0108]
As described above, according to the present embodiment, the imaging optics including the pair of objective optical systems and the image transmission optical system 27 for transmitting an optical image by the objective optical system to the distal end portion 21 of the endoscope 2. Since an image is formed on the image sensor 28 by providing a system, the outer diameter of the insertion section 20 can be reduced, and the first distortion generated by the pair of objective optical systems and the image distortion generated by the image transmission optical system 27. Since the multi-distortion correcting means for correcting the multi-distortion overlapping with the second distortion is provided, accurate measurement can be performed, and the fatigue when stereoscopically viewed can be reduced by correcting the multi-distortion. .
[0109]
In the above description, when the optical data of (a1) to (d) is created, the personal computer 31 is connected to the measurement endoscope apparatus 1 and the operation is performed by the personal computer 31. As a modified example of the embodiment, the measurement endoscope apparatus 1 itself may be used without connecting the personal computer 31.
[0110]
In this case, for example, the CPU 18 a in the measurement processing unit 18 shown in FIG. 2 performs the processes (1-1) to (1-12) in addition to the process of performing the stereo measurement based on the optical data recorded on the memory card 33. ), A process including the function of the multiple distortion correction process 18d and the like. The function of the multiple distortion correction processing 18d by the CPU 18a is shown by a two-dot chain line in FIG.
[0111]
(Second embodiment)
Next, a second embodiment of the present invention will be described with reference to FIGS. The measurement endoscope apparatus of the present embodiment is obtained by changing the configuration of the distal end portion 21 of the endoscope 2 in the measurement endoscope apparatus 1 of FIG.
[0112]
That is, as shown in FIG. 18, the distal end portion 21 of the endoscope 2 includes a distal end main body 39 and an optical adapter 46 detachably attached to the distal end main body 39.
[0113]
The optical adapter 46 is provided with two illumination windows 24 and an observation window 25. FIG. 19 shows the configuration of the distal end portion 21 along the line BB in FIG.
[0114]
As shown in FIG. 19, an image pickup device 28 is attached to the distal end main body 39, and the front side of the image pickup device 28 is protected by a cover glass 47. The cover glass 47 faces a cover glass 48 on the optical adapter 46 side. I do.
[0115]
A fixing ring 49 is provided on the outer peripheral surface on the rear end side of the optical adapter 46, and a male screw portion provided on the outer peripheral surface of the distal end main body 39 is provided on the inner peripheral surface of the rear end endoscope of the fixing ring 49. The female screw portion provided is screwed together so as to be detachably attached.
[0116]
A positioning recess is provided on the outer peripheral surface of the distal end surface of the distal end body 39, and a positioning pin is provided on the optical adapter 39 side. When the optical adapter 39 is attached, positioning in the circumferential direction is performed by the concave portion and the pin. Is done. The objective lenses 26R and 26L and the image transmission optical system 27 are attached to the optical adapter 46 side.
[0117]
Other configurations are basically the same as those of the first embodiment.
[0118]
Next, the operation of the present embodiment will be described.
[0119]
The optical adapter 46 applied to the present measurement endoscope apparatus is attached to a master imaging unit 52 with a built-in imaging element constituting a production measurement jig 51 shown in FIG. Optical data unique to each optical adapter 46 shown in the following (a1) to (d) is measured. The optical data is recorded on a recording medium, for example, a memory card 33. The specific optical data recorded on the memory card 33 corresponds one-to-one with the characteristics of the optical adapter 46, and is handled as one combination after shipment.
[0120]
The above specific optical data is
(A1) Geometric distortion correction table for two objective optical systems
(A2) Geometric distortion correction table of image transmission optical system
(B) The focal length of each of the left and right imaging optical systems
(C) Distance between principal points of left and right imaging optical systems
(D) Optical axis position coordinates on each image of the left and right imaging optical systems
(E) Position information when images of the left and right imaging optical systems are formed on the master image sensor.
[0121]
This embodiment is different from the first embodiment in that an item (e) is added, and the other points are the same as the first embodiment.
[0122]
The optical adapter 46 after taking the above-mentioned specific optical data is attached to the endoscope 2, and the measurement endoscope apparatus performs the following processes (1) to (8) to measure various dimensions. be able to.
[0123]
That is,
(1) The optical data (a1) to (e) is read from the memory card 33. (2) Image a white subject using the present measurement endoscope apparatus.
[0124]
(3) Using the data of (e) and the imaging data of (2), an image position shift due to the combination of the optical adapter 46 and the endoscope 2 is obtained.
[0125]
(4) Using the data of (3) and the data of (1), a conversion table for performing geometric distortion correction on the endoscope 2 is created.
[0126]
(5) The object to be measured, which is a subject, is imaged by the endoscope 2 and an image is captured.
[0127]
(6) The captured image is subjected to coordinate conversion based on the table created in (3).
[0128]
(7) Three-dimensional coordinates of an arbitrary point are obtained by matching the imaging data in the above (2) based on the coordinate-converted image.
[0129]
(8) Various dimensions are measured based on the three-dimensional coordinates.
[0130]
FIG. 20 is a perspective view showing a measurement state by the production measuring jig 51 for measuring optical data specific to the optical adapter 46 described above.
[0131]
The production measuring jig 51 has an optical adapter 46 attachable thereto, and is connected to a master imaging unit 52 having the same structure as the distal end main body 39 of the endoscope 2 and a signal line from the master imaging unit 52. A CCU 53 to be used, a personal computer 31 having a memory card slot 54 in which the memory card 33 is detachable and performing image processing on image data from the CCU 53, and a chart 54 for analyzing optical characteristics of the optical adapter 46. It is configured.
[0132]
When optical data is captured by the production measurement jig 51, first, as shown in FIG. 20, an optical adapter 46 is attached to the master imaging unit 52, and an image of the chart 54 is captured via the optical adapter 46, and the image is obtained. The personal computer 31 performs image processing based on the data, obtains the optical data (a1) to (e), and records the optical data on the memory card 33.
[0133]
Next, each optical data specific to the optical adapter 46 will be specifically described.
[0134]
(A1) (a2) Regarding the geometric distortion correction table,
Generally, an image obtained by a lens system has optical distortion. When performing measurement, this distortion causes a large error. Therefore, this distortion can be removed by performing coordinate transformation. The coordinate conversion may be performed with the center of the optical axis as the center. For more accurate correction, the center of the geometric distortion of the optical system may be used. Also, the geometric distortion correction tables for the two images may be provided separately for the right image and the left image, or the two may be combined into one table. This correction can be performed in the same manner as in the first embodiment.
[0135]
In addition, (b), (c), and (d) are the same as in the first embodiment, and a description thereof will be omitted.
[0136]
(E) Regarding position information when two images are formed on the image sensor of the master, the luminance data of the reference line V (vertical line)
PV (100, Yn) where Yn = 1, 2, 3,... 480
And, as luminance data of the reference line H (horizontal line),
PH (Xn, 100) where Xn = 1, 2, 3,... 640
Measure and record.
[0137]
Next, stereo measurement processing as a measurement operation will be described with reference to the flowcharts of FIGS. 21 and 22 and FIGS.
[0138]
The stereo measurement process is a process of measuring three-dimensional coordinates based on right and left image data having parallax captured by the two objective lenses 26R and 26L of the optical adapter 46.
[0139]
First, in step S30 of FIG. 21, it is checked whether or not to perform initialization. That is, in the initial state after starting the measurement software, and when the optical adapter 46 is replaced or the endoscope is replaced, the process proceeds to the routine of steps S31 to S35, and initialization is performed. Otherwise, the process proceeds to step S11, where the already set corrected image information data is read as original data, and the process proceeds to step S12.
[0140]
In step S31, the correction image data is read. That is, the specific optical data (a1) to (e) corresponding to the optical adapter 46 on a one-to-one basis is read from the memory card 33 and is used as master data.
[0141]
Subsequently, in step S32, an image of a white subject is captured. That is, a white subject is captured by the measurement endoscope apparatus to which the optical adapter 46 is attached, and the CPU 18a of the measurement processing unit 18 loads the image into the RAM 18c via the video signal processing circuit 12.
[0142]
Subsequently, in step S33, a subroutine image position setting process (see FIG. 22A) described later is executed. The process of this subroutine is a process of calculating a shift amount of image coordinates caused by a difference in characteristics between the endoscope 2 and the master imaging unit 52 that has acquired optical data during the production of the optical adapter 46.
[0143]
In step S34, a table is set. In this process, of the optical data fetched in step S31, the correction table obtained by correcting the geometric distortion correction tables of the two types of images (a1) and (a2) in accordance with the displacement of the coordinates obtained in step S33. Is created. That is, original data composed of the following two items is created.
[0144]
As a weight table
W1 '(X, Y) = W1 (X + [Delta] X, Y + [Delta] Y)
W2 '(X, Y) = W2 (X + [Delta] X, Y + [Delta] Y)
W3 '(X, Y) = W3 (X + [Delta] X, Y + [Delta] Y)
W4 '(X, Y) = W4 (X + [Delta] X, Y + [Delta] Y) (9)
Also, as a coordinate reference table,
QX '(X, Y) = QX (X, Y) + [Delta] X
QY ′ (X, Y) = QY (X, Y) + ΔY (10)
However, “with” indicates a state after correction, and “without” indicates a state before correction. Further, ΔX and ΔY indicate the amount of deviation between the image measured by the master and the image when the optical adapter 46 is attached to the apparatus.
[0145]
In step S35, the corrected image information data is stored. That is, the original data obtained in step S34 is stored in the memory card 33.
[0146]
Therefore, in step S12, image data is fetched. That is, an object to be measured is imaged and taken into a memory as original image data. Then, in step S13, an image correction process of a subroutine described later (see FIG. 22B) is executed to perform image correction of the original image. That is, the original image data captured in step S12 is subjected to coordinate conversion based on the original table created in step S34, and measurement image data for performing measurement is obtained.
[0147]
Subsequently, in step S14, based on the measured image data, the above-described subroutine measurement processing (see FIG. 10) is executed, three-dimensional coordinates are calculated, and the selected measurement is performed. finish.
[0148]
Here, the image position setting processing of the subroutine called in step S33 will be described with reference to the flowchart of FIG.
[0149]
First, in step S152, a visual field shape pattern is called. That is, as the luminance data indicating the position information of the image formation of the master (e) with respect to the image pickup device in the specific optical data captured in step S31,
PV (100, Yn), where Yn = 1, 2, 3,... 480
PH (Xn, 100), where Xn = 1, 2, 3,... 640
Call. FIG. 23 is a diagram showing an image obtained by capturing the white chart 54 with the optical adapter 46 attached to the master image capturing unit 52, and the vertical line PV and the horizontal line PH are lines that provide the above-described luminance data, respectively.
[0150]
In step S153, pattern matching, visual field position setting, and optical axis position setting are performed. FIG. 24 is a diagram illustrating an image in which the image of the white chart 54 is captured by the measurement endoscope apparatus to which the optical adapter 46 is attached. As luminance data indicating the position information of the imaging,
PV '(100, Yn), where Yn = 1, 2, 3,... 480
PH '(Xn, 100), where Xn = 1, 2, 3,... 640
Given by
[0151]
FIG. 25 is a diagram showing a change in luminance with the vertical axis representing the luminance that gives the visual field shape pattern to the master and the measurement endoscope apparatus 1 and the horizontal axis representing the address of the image. (A) shows a change in luminance of the line PV and the line PV ′, and FIG. 25 (B) shows a change in luminance of the line PH and the line PH ′.
[0152]
In the pattern matching processing in step S153, the positional deviations ΔX and ΔY of the image with respect to the master are obtained using normalized cross-correlation based on the luminance changes in FIGS. 25A and 25B. Further, the horizontal and vertical coordinates xR ', yR' and xL ', yL' of the measurement system optical axis position with respect to the right and left optical systems are obtained by the following equations. That is,
xR '= XR + [Delta] X
yR '= YR + [Delta] Y
xL '= XL + [Delta] X
yL ′ = YL + ΔY (11)
It becomes.
[0153]
The image correction processing of the subroutine called in step S13 (see FIG. 21) will be described with reference to the flowchart in FIG. The pixel data P (X, Y) on the measurement image can be obtained by the processing of this subroutine. First, in step S162, the original data which is the correction table created in step S34 (see FIG. 21) is fetched. In step S163, the coordinates of the corrected image, that is, the coordinates of the measured image are set as X and Y.
[0154]
In step S164, the coordinates of the original image necessary to obtain the corrected coordinates X and Y are used as reference coordinates, and the corrected coordinates X and Y are obtained by using QX '(X, Y) and QY' (X, Y) created in step S34. Get Y. Then, in step S165, the weight table created in step S34 is fetched, and in step S166, the pixel data P (X, Y) on the coordinates after correction is obtained by the following equation. That is,

Required by
[0155]
Subsequently, in step S167, a corrected image range is set, and this routine ends. Note that the image correction processing of this subroutine is performed on the left and right images, and the original image is converted into a measurement image.
[0156]
The measurement processing of the subroutine called in step S14 of FIG. 21 has been described with reference to FIG.
[0157]
Further, as an example of measurement in the present embodiment, correction may be performed internally only on luminance information and an image may not be created, and a measurement point may be specified on an image before correction.
[0158]
In this case, since the correction image creation processing can be simplified, the time from image capture to measurement start can be reduced.
[0159]
As an example of use of this embodiment, it can be performed as described in the first embodiment, and correction is performed internally only on luminance information as described below. Instead of creating an image, a measurement point may be specified on the image before correction.
[0160]
In this use example, since the process of creating the corrected image can be simplified, the time from image capture to the start of measurement can be reduced.
[0161]
In the present embodiment, the case where the stereo optical adapter 46 is mounted has been described, but another optical adapter can be mounted and used.
[0162]
As described above, according to the measurement endoscope apparatus of the present embodiment, in order to improve measurement accuracy, specific optical data is stored in a memory card of a recording medium so as to correspond to the optical adapter 46 on a one-to-one basis. 33, the accuracy of measurement can be improved by correcting the displacement of the coordinates between the production measurement jig 51 (master imaging unit 52) and the endoscope image.
[0163]
In addition, since specific optical data was collected using the production measurement jig 51 (master imaging device 52) in the production of the measurement optical adapter 46, the productivity was improved and the measurement endoscope device was provided at a low price. become able to.
[0164]
In addition, since the optical adapter 46 for measurement is detachably provided at the distal end portion 21, it can be replaced with an optical adapter other than for other measurement, thereby improving the versatility of the endoscope and reducing the user's financial burden. Can be suppressed.
[0165]
Further, in the example of the present embodiment, the weight tables W1 to W4 and the coordinate reference tables QX and QY used for the image coordinate conversion are created by the computer 31 of the production measuring jig 51 and recorded on the memory card 33 of the recording medium. Therefore, there is no need to create a table in the measurement processing unit inside the endoscope apparatus. Therefore, the measurement process can be performed in a short time.
[0166]
Further, by displaying the value of the normalized cross-correlation on the screen of the LCD 5, the operator can know the reliability of the matching or the reliability of the measurement. When the value of the normalized cross-correlation is small, it is possible to switch to manual matching, and highly accurate measurement can be performed.
[0167]
(Third embodiment)
Next, a third embodiment of the present invention will be described with reference to FIG. In the second embodiment, the objective optical system and the image transmission optical system 27 are all provided on the optical adapter 46 side, and the second distortion is corrected collectively. A part of the transmission optical system 27 is on the optical adapter 46 side, and the second distortion is corrected collectively.
[0168]
FIG. 26 shows the configuration of the distal end portion 21 of the endoscope 2 according to the present embodiment. The distal end portion 21 includes a distal end main body 59 and an optical adapter 60 that is detachably attached to the distal end main body 59. The distal end main body 59 is provided with an image sensor 28 and a rear optical system 27b of the image transmission optical system 27. It is.
[0169]
The optical adapter 60 includes two objective lenses 26R and 26L and an optical system 27a in front of the image transmission optical system 27. The optical adapter 60 is detachably attached to the distal end main body 59 by a fixing ring 49. In this case, the diameter of the luminous flux at the portion divided into the distal end main body 59 and the optical adapter 60, that is, at the surfaces of the cover glasses 61 and 62 becomes large, and the real image is not formed. I keep it small.
[0170]
In this embodiment, the equations (1-1) to (1-4) are used to correct the geometric distortion of the image transmission optical system 27 including the first and second stages, and the equations (1-5) to (1-12) are corrected. Corrects the geometric distortion of the objective optical system.
[0171]
The other configuration and operation are the same as those of the second embodiment. The effects of the present embodiment are almost the same as those of the second embodiment.
[0172]
(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described with reference to FIG. The endoscope device according to the present embodiment is the same as that of the third embodiment.
[0173]
In other words, the distal end portion 21 of the endoscope 2 has the configuration shown in FIG. 26, the upstream optical system 27a of the image transmission optical system 27 is on the optical adapter 60 side, and the rear optical system 27b is on the distal end body 59 side. It is provided.
[0174]
However, in the present embodiment, the second distortion is separately corrected for the former stage and the latter stage.
[0175]
The measurement processing unit 18 has a function of geometric multiple distortion correction processing, and the function of geometric multiple distortion correction processing is a function of a first distortion correction processing and a function of a second distortion pre-stage correction processing. , And the function of the second distortion post-stage correction process.
[0176]
Equations (13-1) to (13-4) shown below correct the geometric distortion at the later stage of the transmission optical system, and equations (13-5) to (13-8) show the geometric distortion at the earlier stage of the image transmission optical system. The geometrical distortion of the objective optical system is corrected by the equations (13-9) to (13-16).
[0177]
x, y: coordinates before correction
x_R’, Y_R’: Right screen coordinate after correction
x_L’, Y_L’: Left screen coordinate after correction
c_Bx, C_By: Center coordinate of the second stage after the second distortion
c_Fx, C_Fy: Center coordinate of the second stage before the second distortion
c_Rx, C_Ry: 1st distortion right center coordinate
c_Lx, C_Ly: 1st distortion left center coordinate
k_x, K_y, A_ij, B_ij: Second distortion correction coefficient
k_Rx, K_Ry, A_Rij, B_Rij: 1st distortion right correction coefficient
k_Lx, K_Ly, A_Lij, B_Lij: 1st distortion left correction coefficient
Second distortion second-stage correction:
f_2Bx(X, y) ≡k_Bx(A_B00+ A_B12xy²+ A_B14xy⁴+ A_B16xy⁶+ A_B30x³+ A_B32x³y²+ A_B34x³y⁴
+ A_B50x⁵+ A_B52x⁵y²+ A_B70x⁷) + C_Bx(13-1)
f_{2 By}(X, y) ≡k_By(B_B00+ B_B21x²y + b_B41x⁴y + b_B61x⁶y + b_B03y³+ B_B23x²y³+ B_B43x⁴y³
+ B_B05y⁵+ B_B25x²y⁵+ B_B07y⁷) + C_By(13-2)
x₂₁= F_2x(Xc_Bx, Y-c_By) (13-3)
y₂₁= F_2y(Xc_Bx, Y-c_By) (13-4)
Second distortion pre-stage correction:
f_2Fx(X, y) ≡k_Fx(A_F00+ A_F12xy²+ A_F14xy⁴+ A_F16xy⁶+ A_F30x³+ A_F32x³y²+ A_F34x³y⁴
+ A_F50x⁵+ A_F52x⁵y²+ A_F70x⁷) + C_Fx(13-5)
f_2Fy(X, y) ≡k_Fy(B_F00+ B_F21x²y + b_F41x⁴y + b_F61x⁶y + b_F03y³+ B_F23x²y³+ B_F43x⁴y³
+ B_F05y⁵+ B_F25x²y⁵+ B_F07y⁷) + C_Fy(13-6)
x₂= F_2x(X₂₁-C_Fx, Y₂₁-C_Fy) (13-7)
y₂= F_2y(X₂₁-C_Fx, Y₂₁-C_Fy) (13-8)
First distortion right correction:
f_Rx(X, y) ≡k_Rx(A_R00+ A_R12xy²+ A_R14xy⁴+ A_R16xy⁶+ A_R30x³+ A_R32x³y²+ A_R34x³y⁴
+ A_R50x⁵+ A_R52x⁵y²+ A_R70x⁷) + C_Rx(13-9)
f_Ry(X, y) ≡k_Ry(B_R00+ B_R21x²y + b_R41x⁴y + b_R61x⁶y + b_R03y³+ B_R23x²y³+ B_R43x⁴y³
+ B_R05y⁵+ B_R25x²y⁵+ B_R07y⁷) + C_Ry(13-10)
X_R= F_Rx(X₂-C_Rx, Y₂-C_Ry) (13-11)
Y_R= F_Ry(X₂-C_Rx, Y₂-C_Ry) (13-12)
First distortion left correction:
f_Lx(X, y) ≡k_Lx(A_L00+ A_L12xy²+ A_L14xy⁴+ A_L16xy⁶+ A_L30x³+ A_L32x³y²+ A_L34x³y⁴
+ A_L50x⁵+ A_L52x⁵y²+ A_L70x⁷) + C_Lx(13-13)
f_Ly(X, y) ≡k_Ly(B_L00+ B_L21x²y + b_L41x⁴y + b_L61x⁶y + b_L03y³+ B_L23x²y³+ B_L43x⁴y³
+ B_L05y⁵+ B_L25x²y⁵+ B_L07y⁷) + C_Ly(13-14)
x_L’= F_Lx(X₂-C_Lx, Y₂-C_Ly) (13-15)
y_L’= F_Ly(X₂-C_Lx, Y₂-C_Ly) (13-16)
According to the endoscope 2 on which the optical adapter 59 is actually mounted, it is possible to cope with a case where the amount of displacement between the center of the second stage of the second stage and the center of the second stage of the second stage is changed, so that more accurate correction can be performed. I have to.
[0178]
The optical data in this case is
(A1) Geometric distortion correction table for two objective optical systems
(A2) Geometric distortion correction table before image transmission optical system
(A3) Geometric distortion correction table after image transmission optical system
(B) The focal length of each of the left and right imaging optical systems
(C) Distance between principal points of left and right imaging optical systems
(D) Optical axis position coordinates on each image of the left and right imaging optical systems
(E) Position information when images of the left and right imaging optical systems are formed on the master image sensor.
[0179]
Referring to the optical data, geometric multiple distortion correction is performed as shown in FIG.
[0180]
That is, when the coordinate conversion is started as shown in FIG. 27, the second post-distortion correction data is read from the second post-distortion correction table in the first step S41, and the rear optical system of the image transmission optical system 27 is read in the next step S42. The second-stage correction of the second distortion of 27b is performed. In the next step S43, the second distortion pre-stage correction data is read from the second distortion pre-stage correction table, and in the next step S44, the second distortion post-stage correction of the pre-stage optical system 27a of the image transmission optical system 27 is performed.
[0181]
In the next step S45, the first distortion correction data is read from the first distortion correction table, and in the next step S46, the first distortion correction of the objective optical system is performed.
[0182]
By performing the correction in this manner, a more accurate correction can be performed according to the endoscope 2 on which the optical adapter 59 is actually mounted. The other effects are the same as those of the third embodiment.
[0183]
(Fifth embodiment)
Next, a fifth embodiment of the present invention will be described. This embodiment is different from the first to third embodiments in that the following equations (14-1) to (14-5) are used instead of the second distortion correction equations of equations (1-1) to (1-4). ). The effect in this case is almost the same as in the first to third embodiments.
[0184]
h₂(X, y) ≡K₂Sin^-1｛(X²+ Y²)^1/2/ (D₂K₂)｝ (14-1)
h₂(X, y) ≡K₂Tan^-1｛(X²+ Y²)^1/2/ (D₂K₂)｝ (14-1 ’)
(14-1) and (14-1 ') are selectively used depending on the characteristic of the geometric distortion of the lens.
[0185]
f_2x(X, y) ≡k_2xd₂Cos ｛tan^-1(Y / x) @ tan @ h₂(X, y)｝ + c_x(14-3)
f_2y(X, y) ≡k_2yd₂Cos ｛tan^-1(Y / x) @ tan @ h₂(X, y)｝ + c_y(14-3)
x₂= F_2x(Xc_x, Y-c_y) (14-4)
y₂= F_2y(Xc_x, Y-c_y) (14-5)
here,
x, y: coordinates before correction
x₂, Y₂: Coordinates after the second distortion correction
c_x, C_y: Second distortion center coordinates
K₂, D₂, K_2x, K_2y: Second distortion correction coefficient
It is.
[0186]
For the fourth embodiment, the following equations (15-1) to (15-10) are used instead of the second distortion correction equations of equations (13-1) to (13-8). Just do it. The effect in this case is almost the same as that of the fourth embodiment.
[0187]
Second distortion second-stage correction:
h_B(X, y) ≡K_BSin^-1｛(X²+ Y²)^1/2/ (D_BK_B)｝ (15-1)
h_B(X, y) ≡K_BTan^-1｛(X²+ Y²)^1/2/ (D_BK_B)｝ (15-1 ’)
(15-1) and (15-1 ') are selectively used depending on the characteristic of the geometric distortion of the lens.
[0188]
f_Bx(X, y) ≡k_Bxd_BCos ｛tan^-1(Y / x) @ tan @ h_B(X, y)｝ + c_Bx(15-2)
f_By(X, y) ≡k_Byd_BCos ｛tan^-1(Y / x) @ tan @ h_B(X, y)｝ + c_By(15-3)
x_B= F_Bx(Xc_Bx, Y-c_By) (15-4)
y_B= F_By(Xc_Bx, Y-c_By) (15-5)
Second distortion pre-stage correction:
h_F(X, y) ≡K_FSin^-1｛(X²+ Y²)^1/2/ (D_FK_F)｝ (15-6)
h_F(X, y) ≡K_FTan^-1｛(X²+ Y²)^1/2/ (D_FK_F)｝ (15-6 ')
(15-6) and (15-6 ') are selectively used depending on the geometric distortion characteristics of the lens.
[0189]
f_Fx(X, y) ≡k_Fxd_FCos ｛tan^-1(Y / x) @ tan @ h_F(X, y)｝ + c_Fx(15-7)
f_Fy(X, y) ≡k_Fyd_FCos ｛tan^-1(Y / x) @ tan @ h_F(X, y)｝ + c_Fy(15-8)
x₂= F_Fx(Xc_Fx, Y-c_Fy) (15-9)
y₂= F_Fy(Xc_Fx, Y-c_Fy) (15-10)
here,
x, y: coordinates before correction
x₂, Y₂: Coordinates after the second distortion correction
c_Fx, C_Fy: Center coordinate of the second stage before the second distortion
c_Bx, C_By: Center coordinate of the second stage after the second distortion
K_F, D_F, K_Fx, K_Fy: Center coordinate of the second stage before the second distortion
K_B, D_B, K_Bx, K_By: Center coordinate of the second stage after the second distortion
In the above-described first to fourth embodiments, the coordinate transformation formula for the first distortion correction may be as follows.
[0190]
First distortion right correction:
h_R(X, y) ≡K_RSin^-1｛(X²+ Y²)^1/2/ (D_RK_R)｝ (16-1)
h_R(X, y) ≡K_RTan^-1｛(X²+ Y²)^1/2/ (D_RK_R)｝ (16-1 ’)
(16-1) and (16-1 ') are selectively used depending on the characteristic of the geometric distortion of the lens.
[0191]
f_Rx(X, y) ≡k_Rd_RCos ｛tan^-1(Y / x) @ tan @ h_R(X, y)｝ + c_Rx(16-2)
f_Ry(X, y) ≡k_Rd_RCos ｛tan^-1(Y / x) @ tan @ h_R(X, y)｝ + c_Ry(16-3)
x_R= F_Rx(Xc_Rx, Y-c_Ry) (16-4)
y_R= F_Ry(Xc_Rx, Y-c_Ry) (16-5)
First distortion left correction:
h_L(X, y) ≡K_LSin^-1｛(X²+ Y²)^1/2/ (D_LK_L)｝ (16-6)
h_L(X, y) ≡K_LTan^-1｛(X²+ Y²)^1/2/ (D_LK_L)｝ (16-6 ')
(16-6) and (16-6 ') are selectively used depending on the characteristic of the geometric distortion of the lens.
[0192]
f_Lx(X, y) ≡k_Ld_LCos ｛tan^-1(Y / x) @ tan @ h_L(X, y)｝ + c_Lx(16-7)
f_Ly(X, y) ≡k_Ld_LCos ｛tan^-1(Y / x) @ tan @ h_L(X, y)｝ + c_Ly(16-8)
x_L= F_Lx(X₂-C_Lx, Y₂-C_Ly) (16-9)
y_L= F_Ly(X₂-C_Lx, Y₂-C_Ly) (16-10)
here,
x₂, Y₂: Coordinates after the second distortion correction
x_R, Y_R: Right screen coordinates after correction
x_L, Y_L: Left screen coordinates after correction
c_Rx, C_Ry: 1st distortion right center coordinate
c_Lx, C_Ly: 1st distortion left center coordinate
K_R, D_R, K_Rx, K_Ry: 1st distortion right correction coefficient
K_L, D_L, K_Lx, K_Ly: 1st distortion left correction coefficient
It is.
[0193]
Note that, as described above, using the optical data for correcting the multiple distortion, for an existing endoscope having a configuration in which an image is formed on an image sensor through a pair of objective optical systems and an image transmission optical system, It can also be applied to correct the image.
[0194]
In this case, by performing multiple distortion correction, for example, by reading the image data through an image processing unit that performs image correction via a recording medium such as a memory card 33 that stores optical data, highly accurate measurement and image display can be performed. It becomes possible. In addition, an operation program for performing the multiple distortion correction using the optical data may be recorded together with the optical data, and the multiple distortion correction may be performed according to the operation program.
[0195]
[Appendix]
1. In claim 3, in the geometric multiple distortion correction, first the second distortion correction is performed, and then the first distortion correction is performed.
[0196]
2. In claim 5, the geometric multiple distortion correction includes first performing a second distortion second-stage correction, then performing a second distortion first-stage correction, and finally performing a first distortion correction.
[0197]
3. A plurality of objective optical systems for obtaining images from different viewpoints, and an output signal from an image sensor that forms a plurality of images by an image transmission optical system that transmits images of the plurality of objective optical systems, Geometric multiple distortion correction means for correcting multiple distortion, which is a distortion in which the first geometric distortion by the plurality of objective optical systems and the second geometric distortion by the image transmission optical system, are provided. Image processing device.
[0198]
4. A plurality of objective optical systems for obtaining images from different viewpoints, and an output signal from an image sensor that forms a plurality of images by an image transmission optical system that transmits images of the plurality of objective optical systems, A recording medium that records data for correcting multiple distortion, which is distortion in which a first geometric distortion caused by the plurality of objective optical systems and a second geometric distortion caused by the image transmission optical system overlap.
[0199]
【The invention's effect】
According to the embodiments of the present invention described above, an imaging optical system including a pair of objective optical systems and an image transmission optical system for transmitting an optical image by the objective optical system is provided at the distal end of the endoscope. Since an image is formed on the element, the outer diameter of the insertion portion can be reduced, and the first distortion generated by the pair of objective optical systems and the second distortion generated by the image transmission optical system overlap. Since the multiple distortion correcting means for correcting multiple distortions is provided, accurate measurement can be performed, and by correcting the multiple distortions, the feeling of fatigue when stereoscopically viewed can be reduced.
[Brief description of the drawings]
FIG. 1 is a perspective view showing an entire measurement endoscope apparatus according to a first embodiment of the present invention.
FIG. 2 is a block diagram showing a configuration of the measurement endoscope apparatus of FIG.
FIG. 3 is an enlarged perspective view showing a distal end portion of an insertion section of the endoscope.
FIG. 4 is a sectional view taken along line AA of FIG. 3;
FIG. 5 is a pixel arrangement diagram of an image before correction of geometric distortion and an image after correction.
FIG. 6 is a diagram showing an image pixel before correction and an image pixel after correction.
FIG. 7 is a flowchart showing the contents of coordinate conversion processing including multiple distortion correction.
FIG. 8 is a flowchart showing the contents of a stereo measurement process.
FIG. 9 is a flowchart showing the contents of a subroutine called in the stereo measurement processing of FIG. 8;
FIG. 10 is a flowchart showing the contents of a subroutine called in the measurement processing of FIG. 8;
FIG. 11 is a perspective view showing a pipe in a depth measurement state using the measurement endoscope apparatus.
FIG. 12 is an enlarged view of a measurement part of the pipe of FIG. 11;
FIG. 13 is a perspective view showing a pipe in a length measurement state using the measurement endoscope device.
FIG. 14 is a perspective view showing the turbine blade in a dent measurement state using the measurement endoscope apparatus.
15 shows a subroutine called in the measurement processing of FIG. 10, FIG. 15 (A) is a flowchart of a three-dimensional coordinate analysis processing, and FIG. 15 (B) is a flowchart of a pattern matching processing.
FIG. 16 is a diagram showing a right and left two image positional relationship on a three-dimensional coordinate system having x, y, and z axes for explaining a basic principle of three-dimensional coordinate analysis processing in the measurement endoscope apparatus.
FIG. 17 is a diagram showing a stereo measurement screen in which two left and right images by the measurement endoscope apparatus are displayed on an LCD.
FIG. 18 is a perspective view showing a distal end portion of an insertion section according to the second embodiment of the present invention.
FIG. 19 is a sectional view taken along line BB of FIG. 18;
FIG. 20 is a perspective view showing a state in which an optical adapter is connected to a master imaging unit of a production measurement jig to collect optical data.
FIG. 21 is a flowchart showing the contents of a stereo measurement process.
FIG. 22 is a flowchart showing the contents of a subroutine called in the stereo measurement processing of FIG. 21;
FIG. 23 is a diagram illustrating an image of a white chart captured by connecting an optical adapter to a master imaging unit.
FIG. 24 is a view showing a captured white chart image.
25A and 25B are luminance waveforms that give a visual field shape pattern to the measurement endoscope apparatus and the master, where FIG. 25A shows a change in luminance of PV and PV ′ in the vertical direction of an image, and FIG. The figure which shows the brightness | luminance change of PH and PH 'of the horizontal direction of an image.
FIG. 26 is a sectional view showing the structure of the distal end portion of the insertion section according to the third embodiment of the present invention.
FIG. 27 is a flowchart showing the contents of processing of coordinate conversion according to the fourth embodiment of the present invention.
[Explanation of symbols]
1. Measurement endoscope device
2. Endoscope
3. Control unit
4: Remote control
5 LCD
6 ... FMD
9 ... CCU
18 Measurement processing unit
18a CPU
20 ... insertion part
21 ... Tip
26R, 26L: Objective optical system
27 ... Image transmission optical system
31 ... PC
33… Memory card

Claims (5)

A plurality of objective optical systems for obtaining images from different viewpoints,
An image transmission optical system for transmitting images of the plurality of objective optical systems,
An imaging element that forms a plurality of images transmitted to the image transmission optical system,
A processing unit that receives a signal from the image sensor and generates a video signal;
Image distortion correction means for correcting optical distortion of the video signal,
In an endoscope device comprising:
The image distortion correction means,
Geometric multiple distortion correction means for correcting multiple distortion, which is distortion in which the first geometric distortion by the plurality of objective optical systems and the second geometric distortion by the image transmission optical system overlap. An endoscope apparatus characterized by the above-mentioned. The geometric multiple distortion correction means,
First distortion correction means for correcting the first distortion by the plurality of objective optical systems for each of the plurality of objective optical systems;
Second distortion correction means for correcting a second distortion caused by the image transmission optical system;
The endoscope apparatus according to claim 1, comprising: The optical system according to claim 1, wherein the objective optical system and the image transmission optical system are configured in an optical adapter, and the optical adapter is configured to be detachable from an insertion unit distal end main body provided with the imaging device. 3. The endoscope device according to 2. The image transmission optical system is divided into an objective optical system side and an image sensor side,
The objective optical system and the objective optical system side of the transmission optical system are configured in an optical adapter, and the optical adapter is configured to be detachable from the image sensor side of the image transmission system. The endoscope device according to claim 1. The second distortion correction means corrects geometric distortion caused by the objective optical system in the transmission system, and the second distortion correction means corrects geometric distortion caused by the image sensor in the image transmission system. The endoscope apparatus according to claim 4, further comprising a post-distortion correction unit.

Images