JP4246320B2 - Electronic endoscope - Google Patents

Description

即ち、ＴＰは一フレーム分の輝度画素信号の総数である。
【００４３】
ステップ６０３では、算出された平均輝度値ＡＩに対応した絞り開度を絞り２０に与えるべく駆動モータ５８が駆動回路６０によって回転駆動させられる。次いで、ステップ６０４では、第１番目のヒストグラムから最小輝度レベルＬ_MIN 及び最大輝度レベルＬ_MAX が求められてシステムコントローラ３０のＲＡＭに格納される。なお、図８には第１番目のヒストグラムの一例が示され、この例から明らかなように、最小輝度レベルＬ_MIN よりも小さい輝度レベルに対応する輝度画素信号は存在せず、また最大輝度レベルＬ_MAX を越える輝度レベルに対応する輝度画素信号は存在しない。
【００４４】
ステップ６０５では、カウンタＮのカウント数が“１”だけカウントアップされ、次いでステップ６０６ではカウンタＮのカウント数が７に等しいか否かが判断される。現段階では、Ｎは７以下、即ちＮ＝１であるので、本ルーチンは一旦終了する。
【００４５】
1/30sec 経過後、本ルーチンは再び実行されるが、このときヒストグラム抽出回路６４には図９に示すような第２番目のヒストグラムが展開されているものとする。現段階では、Ｎ＝１であるので、ステップ６０１からステップ６０７に進み、そこで第１番目の最小輝度レベルＬ_MIN が輝度レベル８以上であるか否かが判断される。もしＬ_MIN ≧８であれば、ステップ６０８に進み、そこで第１番目の最大輝度レベルＬ_MAX が輝度レベル247 以下であるか否かが判断される。もしＬ_MAX ≦247 であれば、ステップ６０９に進む。
【００４６】
ステップ６０９では、第２番目のヒストグラムの特定の範囲、即ち第１番目のヒストグラムの最小輝度レベルＬ_MIN と最大輝度レベルＬ_MAX との間の範囲よりもその両側で８輝度レベル分だけ拡張された範囲（図９で破線矢印で示す範囲）の全データのうちから一つ置きに間引いたデータに基づいて平均輝度レベルＡＩが算出される。即ち、平均輝度レベルＡＩは以下の式によって算出される。
【００４７】
【数２】

なお、［ｘ］はガウス記号であり、ｘを越えない最大整数を表す。
【００４８】
要するに、２つの連続したフレームの輝度画素信号のヒストグラム、即ち第１番目のヒストグラム（図８）及び第２番目のヒストグラム（図９）については一般的には互いに似たものとなる可能性が高いので（というのは、スコープ１０による患部の観察時でのスコープの動きは少ないからである）、前回の最小輝度レベルＬ_MIN と最大輝度レベルＬ_MAX との間の範囲をそれぞれの境界側でヒストグラム定義域（０≦Ｘ≦255)の例えば３ないし５％分だけ拡張すれば（本実施形態では８輝度レベル分）、その拡大範囲内に第２番目のヒストグラムの実体部分が含まれ得ると仮定することができる。従って、第２番目のヒストグラムについての平均輝度レベルＡＩを算出する際の演算範囲はＬ_MIN −８≦Ｘ≦Ｌ_MAX ＋８と狭まり、しかもその演算は一つ置きの間引きデータに基づくものとなるので、平均輝度レベルＡＩの算出時間は大幅に短縮され得る。
【００４９】
ステップ６０９で第２番目のヒストグラムの平均輝度レベルＡＩが算出されると、ステップ６０９からステップ６０３に進み、そこで該平均輝度値ＡＩに対応した絞り開度を絞り２０に与えるべく駆動モータ５８が駆動回路６０によって回転駆動させられる。次いで、ステップ６０４では、第２番目のヒストグラムから最小輝度レベルＬ_MIN 及び最大輝度レベルＬ_MAX が求められてシステムコントローラ３０のＲＡＭに格納される。
【００５０】
ステップ６０５では、カウンタＮのカウント数が“１”だけカウントアップされ、次いでステップ６０６ではカウンタＮのカウント数が７に等しいか否かが判断される。現段階でも、Ｎは７以下、即ちＮ＝２であるので、本ルーチンは一旦終了する。
【００５１】
ステップ６０７でもしＬ_MIN ＜８であれば、即ち第１番目のヒストグラムが図１０に示すようなものであれば、ステップ６０７からステップ６１０に進み、そこで第１番目の最大輝度レベルＬ_MAX が輝度レベル247 以下であるか否かが判断される。もしＬ_MAX ≦247 であれば、ステップ６１１に進む。なお、上述した場合と同様な理由により、第１番目のヒストグラムが図１０に示すようなものであれば、第２番目のヒストグラムはそれと類似したもの、即ち図１１に示すようなものと仮定し得る。
【００５２】
ステップ６１１では、図１１に示す第２番目のヒストグラムの特定の範囲、即ち第１番目のヒストグラムの最小輝度レベルＬ_MIN と最大輝度レベルＬ_MAX との間の範囲よりも最小輝度レベルＬ_MIN 側で輝度レベル０まで拡張されかつ最大輝度レベルＬ_MAX 側で８輝度レベル分だけ拡張された範囲（図１１で破線矢印で示す範囲）の全データのうちから一つ置きに間引いたデータに基づいて平均輝度レベルＡＩが算出される。即ち、平均輝度レベルＡＩは以下の式によって算出される。
【００５３】
【数３】

【００５４】
上述の場合と同様に、第２番目のヒストグラムについての平均輝度レベルＡＩを算出する際の演算範囲は０≦Ｘ≦Ｌ_MAX ＋８と狭まり、しかもその演算は一つ置きの間引きデータに基づくものとなるので、平均輝度レベルＡＩの算出時間は大幅に短縮され得る。
【００５５】
ステップ６１１で第２番目のヒストグラムの平均輝度レベルＡＩが算出されると、ステップ６１１からステップ６０３に進み、そこで該平均輝度値ＡＩに対応した絞り開度を絞り２０に与えるべく駆動モータ５８が駆動回路６０によって回転駆動させられる。次いで、ステップ６０４では、第２番目のヒストグラムから最小輝度レベルＬ_MIN 及び最大輝度レベルＬ_MAX が求められてシステムコントローラ３０のＲＡＭに格納される。
【００５６】
ステップ６０５では、カウンタＮのカウント数が“１”だけカウントアップされ、次いでステップ６０６ではカウンタＮのカウント数が７に等しいか否かが判断される。現段階では、Ｎ＝２であるので、本ルーチンは一旦終了する。
【００５７】
ステップ６０８でもしＬ_MAX ＞247 であれば、即ち第１番目のヒストグラムが図１２に示すようなものであれば、ステップ６０８からステップ６１２に進む。なお、上述した場合と同様な理由により、第１番目のヒストグラムが図１２に示すようなものであれば、第２番目のヒストグラムはそれと類似したもの、即ち図１３に示すようなものと仮定し得る。
【００５８】
ステップ６１２では、図１３に示す第２番目のヒストグラムの特定の範囲、即ち第１番目のヒストグラムの最小輝度レベルＬ_MIN と最大輝度レベルＬ_MAX との間の範囲よりも最小輝度レベルＬ_MIN 側で８輝度レベル分だけ拡張されかつ最大輝度レベルＬ_MAX 側で輝度レベル255 まで拡張された範囲（図１３で破線矢印で示す範囲）の全データのうちから一つ置きに間引いたデータに基づいて平均輝度レベルＡＩが算出される。即ち、平均輝度レベルＡＩは以下の式によって算出される。
【００５９】
【数４】

【００６０】
上述の場合と同様に、第２番目のヒストグラムについての平均輝度レベルＡＩを算出する際の演算範囲はＬ_MIN −８≦Ｘ≦255 と狭まり、しかもその演算は一つ置きの間引きデータに基づくものとなるので、平均輝度レベルＡＩの算出時間は大幅に短縮され得る。
【００６１】
ステップ６１２で第２番目のヒストグラムの平均輝度レベルＡＩが算出されると、ステップ６１２からステップ６０３に進み、そこで該平均輝度値ＡＩに対応した絞り開度を絞り２０に与えるべく駆動モータ５８が駆動回路６０によって回転駆動させられる。次いで、ステップ６０４では、第２番目のヒストグラムから最小輝度レベルＬ_MIN 及び最大輝度レベルＬ_MAX が求められてシステムコントローラ３０のＲＡＭに格納される。
【００６２】
ステップ６０５では、カウンタＮのカウント数が“１”だけカウントアップされ、次いでステップ６０６ではカウンタＮのカウント数が７に等しいか否かが判断される。現段階では、Ｎ＝２であるので、本ルーチンは一旦終了する。
【００６３】
ステップ６１０でもしＬ_MAX ＞247 であれば、即ち第１番目のヒストグラムが図１４に示すようなものであれば、ステップ６１０からステップ６１３に進む。なお、上述した場合と同様な理由により、第１番目のヒストグラムが図１４に示すようなものであれば、第２番目のヒストグラムはそれと類似したもの、即ち図１５に示すようなものと仮定し得る。
【００６４】
ステップ６１３では、図１５に示す第２番目のヒストグラムの全定義域の範囲（図１５で破線矢印で示す範囲、即ち０≦Ｘ≦255 ）の全データのうちから一つ置きに間引いたデータに基づいて平均輝度レベルＡＩが算出される。即ち、平均輝度レベルＡＩは以下の式によって算出される。
【００６５】
【数５】

【００６６】
この場合には、第２番目のヒストグラムについての平均輝度レベルＡＩを算出する際の演算範囲は全定義域（０≦Ｘ≦255 ）となるが、しかしその演算は一つ置きの間引きデータに基づくものとなるので、平均輝度レベルＡＩの算出時間は大幅に短縮され得る。
【００６７】
ステップ６１３で第２番目のヒストグラムの平均輝度レベルＡＩが算出されると、ステップ６１３からステップ６０３に進み、そこで該平均輝度値ＡＩに対応した絞り開度を絞り２０に与えるべく駆動モータ５８が駆動回路６０によって回転駆動させられる。次いで、ステップ６０４では、第２番目のヒストグラムから最小輝度レベルＬ_MIN 及び最大輝度レベルＬ_MAX が求められてシステムコントローラ３０のＲＡＭに格納される。
【００６８】
ステップ６０５では、カウンタＮのカウント数が“１”だけカウントアップされ、次いでステップ６０６ではカウンタＮのカウント数が７に等しいか否かが判断される。現段階では、Ｎ＝２であるので、本ルーチンは一旦終了する。
【００６９】
続いて、第３番目ないし第８番目のヒストグラムについても以上で述べたいずれかの態様（ステップ６０９、６１１、６１２、６１３）でその平均輝度ＡＩが算出され、それに応じて絞り２０の開度が適宜調整される。第８番目のヒストグラムから得られた平均輝度ＡＩに基づく絞り２０の開度の調整後、カウンタＮのカウント数は７に到達する（ステップ６０５）。従って、ステップ６０６からステップ６１４に進み、そこでカウンタＮはリセットされ、本ルーチンは一旦終了する。
【００７０】
1/30sec 経過後、本ルーチンは再び実行されるが、このときＮ＝０となっているので、ステップ６０２に進み、ヒストグラム抽出回路６４で展開された第９番目のヒストグラムの全データ（０≦Ｘ≦255)に基づいて平均輝度値ＡＩが算出され、このとき用いられる演算式は上述の数式１となる。
【００７１】
このとき、ヒストグラムの全データを読み出した上での、最小輝度レベルと最大輝度レベルが求められることになる。従って、例えば、スコープ１０の動きにより内視鏡像の明るさが急激に変化して画像データの存在する範囲が急に異なった場合、或いは画像データの存在する範囲が広がったり狭まったりした場合でも、八回目毎に平均輝度値ＡＩの算出にはヒストグラムの全データ（０≦Ｘ≦255)が用いられるので、適正な平均輝度値が得られ、かくして絞り２０の開度も適正に調整され得る。
【００７２】
図１６ないし図１８を参照すると、システムコントローラ３０で実行されるホワイトバランス設定ルーチンのフローチャートが示される。このホワイトバランス設定ルーチンも電子内視鏡のメイン作動ルーチンのサブルーチンとして機能するものであって、例えば電子内視鏡で映像再現方式としてＮＴＳＣ方式が採用されている場合には、1/30sec 毎に実行される割込みルーチンとされる。また、ホワイトバランス設定ルーチンの実行開始は電源スイッチ７４をオンした後にモード切換スイッチ７８でホワイトバランス設定モードを選択することによって行われる。
【００７３】
なお、ホワイトバランスの設定については、例えば本発明による電子内視鏡が複数のスコープ（１０）と共に病院等の医療機関等に新規に納入されたような場合、本発明による電子内視鏡の納入後に新たなスコープが追加されたような場合、更には既に使用されているスコープについてホワイトバランスの補正係数データを再設定するような場合に行われる。
【００７４】
ステップ１６０１ないしステップ１６０４では、フラグＦ１、Ｆ２、Ｆ３及びＦ４が“０”であるか否かが順次判断される。初期段階では、Ｆ１、Ｆ２、Ｆ３及びＦ４＝０であるので、ステップ１６０５に進み、そこでスイッチ回路６６の接続がフレームメモリ３６側に切り換えられる。従って、ヒストグラム抽出回路６４はスイッチ回路６６を介してフレームメモリ３６に接続される。
【００７５】
ステップ１６０６では、点灯スイッチ７６のオンにより光源ランプ１８が点灯されたか否かが判断される。光源ランプ１８が点灯されていないとき、本ルーチンは一旦終了する。1/30sec 経過後、本ルーチンは再び実行されるが、しかし光源ランプ１８が点灯されない限り、何等の進展もない。
【００７６】
ステップ１６０６で光源ランプ１８の点灯が確認されると、ステップ１６０７に進み、そこで光源ランプ１８を点灯した後に所定の時間が経過したか否かが判断される。本ルーチンの実行は1/30sec 毎に繰り返されるが、しかし所定時間が経過するまでは、何等の進展もない。即ち、光源ランプ１８の発光状態が安定するまで待機状態となる。所定時間が経過して光源ランプ１８の発光状態が安定すると、ステップ１６０８に進み、そこでホワイトバランスの設定が可能である旨のメッセージ、例えば「ホワイトバランスの設定が可能です」というようなメッセージがＴＶモニタ装置４０に表示される。なお、かかるメッセージの表示がシステムコントローラ３０のＲＯＭ内に予め格納された固定文字情報コードデータを読み出してキャラクタ処理回路６８のビデオＲＡＭに書き込んで文字パターン信号としてビデオプロセス回路３８に出力することにより行われることは先に述べた通りである。
【００７７】
ステップ１６０９では、ホワイトバランス設定の対象となるスコープ（１０）が画像信号処理ユニット１２に接続されたか否かが判断される。スコープが接続されていないときは、ステップ１６１０に進み、そこでフラグＦ４が“０”から“１”に書き替えられた後、本ルーチンは一旦終了する。1/30sec 経過後、本ルーチンは再び実行されるが、このときＦ４＝１であるので、ステップ１６０５ないしステップ１６０８はスキップされる。その後、本ルーチンの実行は1/30sec 毎に繰り返されるが、スコープが画像信号処理ユニット１２に接続されない限り、何等の進展もない。
【００７８】
ステップ１６０９でスコープの接続が確認されると、ステップ１６０９からステップ１６１１に進み、そこで該スコープのＥＥＰＲＯＭ７０から補正係数データＧ_GAIN、Ｂ_GAIN及びＲ_GAINが読み出され、これら補正係数データはシステムコントローラ３０のＲＡＭに格納される。次いで、ステップ１６１２では、フラグＦ３が“０”から“１”に書き替えられ、続いてステップ１６１３でホワイトバランス設定スイッチ８０がオンされたか否かが判断される。ホワイトバランス設定スイッチ８０はホワイトバランスの設定準備の完了後にオペレータの手動操作によってオンされる。
【００７９】
ここで、ホワイトバランスの設定準備について簡単に説明すると、ホワイトバランス設定時には基準白色を持つ包囲体が用いられる。即ち、該包囲体の内側壁には所定の基準白色が塗布されており、その包囲体内にスコープ（１０）の先端を挿入することにより、ホワイトバランスの設定準備が完了することになる。なお、電子内視鏡でのホワイトバランスの設定自体は周知のことである。
【００８０】
ステップ１６１３でホワイトバランス設定スイッチ８０のオンが確認されると、ステップ１６１４に進み、そこで撮像センサ１４から読み出された一フレーム分の緑色画素信号に基づいてヒストグラム抽出回路６４で展開されている緑色ヒストグラム（Ｇ）の全データから緑色画素信号の平均値ＡＧが以下の数式に基づいて算出される。
【００８１】
【数６】

ここで、ＧＬ_n は一フレーム分の緑色画素信号についての256 通りの緑色レベルを示し、ＧＳ_n は各緑色レベルに対応した値を持つ緑色画素信号の度数即ち個数を示し、ＴＰは上述した数式１の場合と同様に一フレーム分の緑色画素信号の総数を示す。なお、かかる一フレーム分の緑色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＧ_GAINによってホワイトバランス処理されたものとなっている。
【００８２】
ステップ１６１５では、撮像センサ１４から読み出された一フレーム分の青色画素信号に基づいてヒストグラム抽出回路６４で展開されている青色ヒストグラム（Ｂ）の全データから青色画素信号の平均値ＡＢが以下の数式に基づいて算出される。
【００８３】
【数７】

ここで、ＢＬ_n は一フレーム分の青色画素信号についての256 通りの青色レベルを示し、ＢＳ_n は各青色レベルに対応した値を持つ青色画素信号の度数即ち個数を示し、ＴＰは上述の場合と同様に一フレーム分の青色画素信号の総数を示す。なお、かかる一フレーム分の青色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＢ_GAINによってホワイトバランス処理されたものとなっている。
【００８４】
ステップ１６１６では、平均値ＡＧと平均値ＡＢとの差ΔＧＢが演算される。次いで、ステップ１６１７では、差ΔＧＢが所定の許容値ＰＶを越えているか否かが判断される。差ΔＧＢが許容値ＰＶを越えているとき、ステップ１６１８に進み、そこで平均値ＡＢが平均値ＡＧよりも大きいか否かが判断される。もしＡＢ＞ＡＧであるならば、ステップ１６１９に進み、そこで補正係数データＢ_GAINが所定値だけ減少させられ、本ルーチンは一旦終了する。一方、ＡＢ＜ＡＧであるならば、ステップ１６２０に進み、そこで補正係数データＢ_GAINが所定値だけ増大させられ、本ルーチンは一旦終了させられる。
【００８５】
次の割込み処理で再び本ルーチンが実行されたとき、ステップ１６０１及び１６０２を経た後ステップ１６０３からステップ１６１３まで直ちにスキップすることになるが（Ｆ３＝１）、このときホワイトバランス設定スイッチ８０は既にオンとなっているので、ステップ１６１４に進み、そこで再びヒストグラム抽出回路６４で展開されている緑色ヒストグラム（Ｇ）から緑色画素信号の平均値ＡＧが上述の数式６に基づいて算出される。なお、このとき緑色ヒストグラムは撮像センサ１４から読み出された前回の一フレーム分の緑色画素信号の次に読み出された一フレーム分の緑画素信号に基づくものであり、しかもその後者の一フレーム分の緑色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＧ_GAINによってホワイトバランス処理されたものとなっている。
【００８６】
続いて、ステップ１６１５でも、再びヒストグラム抽出回路６４で展開されている青色ヒストグラム（Ｂ）から青色画素信号の平均値ＡＢが上述の数式７に基づいて算出される。なお、このとき青色ヒストグラムは撮像センサ１４から読み出された前回の一フレーム分の青色画素信号の次に読み出された一フレーム分の青画素信号に基づくものであり、しかもその後者の一フレーム分の青色画素信号は前回の本ルーチン実行時にステップ１６１９或いはステップ１６２０で所定値だけ減少或いは増大させられた補正係数データＢ_GAINによってホワイトバランス処理されたものとなっている。
【００８７】
ステップ１６１６では、平均値ＡＧと平均値ＡＢとの差ΔＧＢが再び演算されるが、その差ΔＧＢは前回よりも小さなものとなる。というのは、今回の一フレーム分の青色画素信号のホワイトバランス処理が前回の本ルーチン実行時にステップ１６１９或いはステップ１６２０で所定値だけ減少或いは増大させられた補正係数データＢ_GAINに基づいて行われているからである。次いで、ステップ１６１７では、差ΔＧＢが所定の許容値ＰＶを越えているか否かが再び判断される。差ΔＧＢが許容値ＰＶを未だ越えているとき、ステップ１６１８に進み、そこで平均値ＡＢが平均値ＡＧよりも大きいか否かが再び判断されるが、このときＡＢ＞ＡＧであれば、ステップ１６１９に進み、そこで補正係数データＢ_GAINが所定値だけ更に減少させられ、またＡＢ＜ＡＧであれば、ステップ１６２０に進み、そこで補正係数データＢ_GAINが所定値だけ更に増大させられる。要するに、補正係数データＧ_GAINを基準にして、差ΔＧＢが所定の許容値ＰＶ内に納まるまで、以上で述べたルーチンが繰り返される。
【００８８】
ステップ１６１７で差ΔＧＢが所定の許容値ＰＶ内に納まったとき、ステップ１６１７からステップ１６２１に進み、そこでフラグＦ２が“０”から“１”に書き替えられる。次いで、ステップ１６２２では、撮像センサ１４から読み出された一フレーム分の緑色画素信号に基づいてヒストグラム抽出回路６４で展開されている緑色ヒストグラム（Ｇ）の全データから緑色画素信号の平均値ＡＧが上述の数式６に基づいて算出される。なお、上述の場合と同様に、かかる一フレーム分の緑色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＧ_GAINによってホワイトバランス処理されたものとなっている。
【００８９】
ステップ１６２３では、撮像センサ１４から読み出された一フレーム分の赤色画素信号に基づいてヒストグラム抽出回路６４で展開されている赤色ヒストグラム（Ｒ）の全データから赤色画素信号の平均値ＡＲが以下の数式に基づいて算出される。
【００９０】
【数８】

ここで、ＲＬ_n は一フレーム分の赤色画素信号についての256 通りの赤色輝度レベルを示し、ＲＳ_n は各赤色レベルに対応した値を持つ赤色画素信号の度数即ち個数を示し、ＴＰは先に述べた場合と同様に一フレーム分の赤色画素信号の総数を示す。なお、かかる一フレーム分の赤色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＲ_GAINによってホワイトバランス処理されたものとなっている。
【００９１】
ステップ１６２４では、平均値ＡＧと平均値ＡＲとの差ΔＧＲが演算される。次いで、ステップ１６２５では、差ΔＧＲが所定の許容値ＰＶを越えているか否かが判断される。差ΔＧＲが許容値ＰＶを越えているとき、ステップ１６２６に進み、そこで平均値ＡＲが平均値ＡＧよりも大きいか否かが判断される。もしＡＲ＞ＡＧであるならば、ステップ１６２７に進み、そこで補正係数データＲ_GAINが所定値だけ減少させられ、本ルーチンは一旦終了する。一方、ＡＲ＜ＡＧであるならば、ステップ１６２８に進み、そこで補正係数データＲ_GAINが所定値だけ増大させられ、本ルーチンは一旦終了させられる。
【００９２】
次の割込み処理で再び本ルーチンが実行されたとき、ステップ１６０１を経た後ステップ１６０２からステップ１６２２まで直ちにスキップし（Ｆ２＝１）、そこで再びヒストグラム抽出回路６４で展開されている緑色ヒストグラム（Ｇ）から緑色画素信号の平均値ＡＧが上述の数式６に基づいて算出される。なお、このとき緑色ヒストグラムは撮像センサ１４から読み出された前回の一フレーム分の緑色画素信号の次に読み出された一フレーム分の緑画素信号に基づくものであり、しかもその後者の一フレーム分の緑色画素信号はスコープのＥＥＰＲＯＭ７０から得られた補正係数データＧ_GAINによってホワイトバランス処理されたものとなっている。
【００９３】
続いて、ステップ１６２３でも、再びヒストグラム抽出回路６４で展開されている赤色ヒストグラム（Ｒ）から赤色画素信号の平均値ＡＲが上述の数式８に基づいて算出される。なお、このとき赤色ヒストグラムは撮像センサ１４から読み出された前回の一フレーム分の赤色画素信号の次に読み出された一フレーム分の赤画素信号に基づくものであり、しかもその後者の一フレーム分の赤色画素信号は前回の本ルーチン実行時にステップ１６２７或いはステップ１６２８で所定値だけ減少或いは増大させられた補正係数データＲ_GAINによってホワイトバランス処理されたものとなっている。
【００９４】
ステップ１６２４では、平均値ＡＧと平均値ＡＲとの差ΔＧＲが再び演算されるが、その差ΔＧＲは前回よりも小さなものとなる。というのは、今回の一フレーム分の赤色画素信号のホワイトバランス処理が前回の本ルーチン実行時にステップ１６２７或いはステップ１６２８で所定値だけ減少或いは増大させられた補正係数データＲ_GAINに基づいて行われているからである。次いで、ステップ１６２５では、差ΔＧＲが所定の許容値ＰＶを越えているか否かが再び判断される。差ΔＧＲが許容値ＰＶを未だ越えているとき、ステップ１６２６に進み、そこで平均値ＡＲが平均値ＡＧよりも大きいか否かが再び判断されるが、このときＡＲ＞ＡＧであれば、ステップ１６２７に進み、そこで補正係数データＲ_GAINが所定値だけ更に減少させられ、またＡＲ＜ＡＧであれば、ステップ１６２８に進み、そこで補正係数データＲ_GAINが所定値だけ更に増大させられる。要するに、補正係数データＧ_GAINを基準にして、差ΔＧＲが所定の許容値ＰＶ内に納まるまで、以上で述べたルーチンが繰り返される。
【００９５】
ステップ１６２５で差ΔＧＲが所定の許容値ＰＶ内に納まったとき、ステップ１６２５からステップ１６２９に進み、そこでホワイトバランス設定が完了した旨のメッセージ、例えば「ホワイトバランス設定が完了しました」というようなメッセージがＴＶモニタ装置４０に表示される。なお、かかるメッセージの表示がシステムコントローラ３０のＲＯＭ内に予め格納された固定文字情報コードデータを読み出してキャラクタ処理回路６８のビデオＲＡＭに書き込んで文字パターン信号としてビデオプロセス回路３８に出力することにより行われることは先に述べた通りである。
【００９６】
ステップ１６３０では、補正係数データＧ_GAIN、Ｂ_GAIN及びＲ_GAINがスコープ（１０）のＥＥＰＲＯＭ７０に書き込まれる。なお、このとき補正係数データＧ_GAINについては変えられていないので、補正係数データＢ_GAIN及びＲ_GAINだけがＥＥＰＲＯＭ７０に書き込まれてもよい。次いで、ステップ１６３１では、当該スコープ（１０）、即ちホワイトバランスの設定或いは再設定が完了したスコープが画像信号処理ユニット１２から離脱させられたか否かが判断される。当該スコープ（１０）の離脱が行われていないとき、ステップ１６３２に進み、そこでフラグＦ１が“０”から“１”に書き替えられる。その後、1/30sec 経過毎に本ルーチンは実行されるが、当該スコープ（１０）が離脱されるまで、何等の進展もない。ステップ１６３１で当該スコープ（１０）の離脱が確認されると、ステップ１６３３に進み、そこでフラグＦ１、フラグＦ２及びフラグＦ３がそれぞれ“１”から“０”に書き直され、本ルーチンは一旦完了する。
【００９７】
その後、ステップ１６０９において、別のスコープ、即ちホワイトバランスの設定或いは再設定されるべきスコープが画像信号処理ユニット１２に接続されたか否かが監視され、もしそのようなスコープが接続されると、以上で述べたルーチンが再び繰り返される。なお、モード切換スイッチ７８が通常の作動モードを選択するように切り換えられると、本ルーチンは終了する。
【００９８】
以上で説明した実施形態では、電子内視鏡は面順次カラー方式を採用したものとされているが、しかしカラー同時方式を採用した電子内視鏡、即ち微細な三原色フィルタ要素をモザイク状に配列したカラーフィルタアレイを固体撮像素子の受光面に適用した電子内視鏡についても同様な自動調光及びホワイトバランス設定を行い得ることが理解されるべきである。更に、以上の説明では、ホワイトバランス処理について、三原色（Ｒ、Ｇ、Ｂ）のそれぞれの色のデータから、三原色の色の補正係数データを設定するようにしているが、三原色の色の補正係数データを求めるために色のデータではなく色差信号データを用いてもよい。
【００９９】
また、上述の実施形態では、ヒストグラムの作成については一フレーム分の画素信号に基づいているが、一フィールド分の画素信号に基づいてヒストグラムの作成を行うようにしてもよい。
【０１００】
【発明の効果】
以上の記載から明らかなように、本発明による電子内視鏡にあっては、一フレーム分の輝度画素信号についてヒストグラム抽出回路で展開されたヒストグラムに基づいて絞り開度を制御する際に該ヒストグラムからの間引きデータから平均輝度値を算出し、その平均輝度値に基づいて絞り開度を制御して自動調光を行うので、その自動調光制御の応答性を高めて再現画像の輝度の適正化を迅速に行うことが可能となる。また、本発明による電子内視鏡にあっては、かかる絞り開度の制御に用いられるヒストグラム抽出回路を利用してホワイトバランス設定を適宜行うことができ、この場合はヒストグラムデータを間引かず全データを用いるので、個々のスコープから得られる画素信号に対して常に適正なホワイトバランス処理を施すことが可能であり、電子内視鏡による内視鏡像の色再現性を高品位に維持することができる。
【図面の簡単な説明】
【図１】本発明による電子内視鏡の概略ブロック図である。
【図２】図１に示す回転式ＲＧＢカラーフィルタの正面図である。
【図３】図１に示す回転式ＲＧＢカラーフィルタをその駆動モータと共に示す側面図である。
【図４】図１に示す絞りをその駆動機構と共に示す概略正面図である。
【図５】図１に示すヒストグラム抽出回路で展開されるヒストグラムの一例を概念的に示すグラフである。
【図６】図１のシステムコントローラで実行される自動調光ルーチンを説明するためのフローチャートの一部分である。
【図７】図１のシステムコントローラで実行される自動調光ルーチンを説明するためのフローチャートの残りの部分である。
【図８】一フレーム分の輝度画素信号から得られるヒストグラムの一例を概念的に示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図９】図８のヒストグラムに関連したヒストグラムを示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１０】一フレーム分の輝度画素信号から得られるヒストグラムの別の例を概念的に示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１１】図１０のヒストグラムに関連したヒストグラムを示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１２】一フレーム分の輝度画素信号から得られるヒストグラムの更に別の例を概念的に示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１３】図１２のヒストグラムに関連したヒストグラムを示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１４】一フレーム分の輝度画素信号から得られるヒストグラムの更に別の例を概念的に示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１５】図１４のヒストグラムに関連したヒストグラムを示すグラフであって、図６及び図７の自動調光ルーチンを説明するための説明図である。
【図１６】図１のシステムコントローラで実行されるホワイトバランス設定ルーチンを説明するためのフローチャートの一部分である。
【図１７】図１のシステムコントローラで実行されるホワイトバランス設定ルーチンを説明するためのフローチャートのその他の部分である。
【図１８】図１のシステムコントローラで実行されるホワイトバランス設定ルーチンを説明するためのフローチャートの残りの部分である。
【符号の説明】
１０ スコープ
１２ 画像信号処理ユニット
１４ 撮像センサ
１６ 光ガイド
１８ 白色光源
２０ 絞り
２２ 集光レンズ
２４ 回転式ＲＧＢカラーフィルタ
２８ ＣＣＤドライバ
３０ システムコントローラ
３２ ＣＣＤプロセス回路
３４ アナログ／デジタル（Ａ／Ｄ）変換器
３６ フレームメモリ
３８ ビデオプロセス回路
４０ ＴＶモニタ装置
６４ ヒストグラム抽出回路
６６ スイッチ回路
７０ ＥＥＰＲＯＭ
７４ 電源スイッチ
７６ 点灯スイッチ
７８ モード切換スイッチ
８０ ホワイトバランス設定スイッチ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an electronic endoscope including a scope made of a flexible conduit and an image signal processing unit for detachably connecting the scope.
[0002]
[Prior art]
In an electronic endoscope of the type described above, a solid-state imaging device, for example, a CCD (charge coupled device) image sensor is provided at the distal end of the scope, and this CCD image sensor is combined with an objective lens system. In addition, an illumination light guide composed of an optical fiber bundle is inserted into the scope, and the end surface of the distal end thereof is combined with the illumination lens.
[0003]
An illumination light source such as a halogen lamp or a xenon lamp is provided in the image signal processing unit, and the proximal end of the illumination light guide is optically connected to the illumination light source when the scope and the image signal processing unit are connected. Thus, when the scope is inserted into the body cavity of the patient, the front of the objective lens system at the distal end is illuminated with the illumination light emitted from the end face of the distal end of the illumination light guide of the scope, thereby the optical subject. The image is formed on the light receiving surface of the CCD image sensor, where it is photoelectrically converted as a pixel signal. The pixel signal obtained by the CCD image sensor is sent to an image signal processing unit, where a video signal is created based on the pixel signal. The video signal is then output from the image signal processing unit to the TV monitor device, where an optical subject image is reproduced on the TV monitor device.
[0004]
By the way, generally, the depth of focus of the objective lens system of the scope of the electronic endoscope is relatively deep. This is because it is necessary not only to observe the distal end of the scope close to the affected area such as a lesion, but also to find the affected area such as a lesion, to observe the entire wide area including the affected area. is there. In this case, in order to always reproduce the optical subject image with appropriate luminance, it is necessary to appropriately adjust the amount of light emitted from the distal end of the illumination light guide. That is, when observing with the distal end of the scope closest to the affected part such as a lesion, the amount of light is reduced to the lowest level, and the distal end of the scope gradually moves away from the affected part to gradually increase the amount of light. is necessary.
[0005]
The light amount adjustment as described above is called automatic dimming. In general, for this automatic dimming, the aperture of the diaphragm incorporated in the light source is controlled based on the average luminance level of the pixel signal for one frame. Is done. More specifically, the histogram extraction circuit develops a histogram for the pixel signal for one frame, the average luminance level of the pixel signal for one frame is calculated from this histogram, and the aperture opening is controlled based on the calculation result. Thus, the brightness of the reproduced image can be properly maintained.
[0006]
[Problems to be solved by the invention]
By the way, when the PAL system is adopted as the TV video reproduction system, the calculation of the average luminance level of the pixel signal for one frame must be repeated every 1/25 sec. When the method is adopted, the calculation of the average luminance level of the pixel signal for one frame must be repeated every 1/30 sec.
[0007]
Therefore, in order to improve the responsiveness of the aperture opening control to the average luminance level of the pixel signal for one frame as much as possible, that is, to quickly optimize the overall luminance level of the reproduced image, the average luminance is required. It is necessary to perform level calculation as quickly as possible. However, conventionally, since the calculation of the average luminance level is intended for all the pixel signals of the histogram, that is, all the pixel signals for one frame, the calculation takes a relatively long time, and the reproduced image It has been pointed out as a problem that it is not possible to quickly optimize the brightness of the screen.
[0008]
Accordingly, an object of the present invention is to quickly optimize the brightness of a reproduced image by increasing the responsiveness as much as possible when controlling the aperture opening based on the histogram developed by the histogram extraction circuit for the pixel signal for one frame. It is an object of the present invention to provide an electronic endoscope that can be performed at the same time.
[0009]
On the other hand, in the conventional electronic endoscope system as described above, a problem that white balance processing cannot be performed properly has been pointed out. In detail, when each scope is shipped from the factory, each scope is connected to a reference image signal processing unit called a master image signal processing unit, in which correction coefficients for white balance processing are set, and the correction is performed. The coefficient data is written in a nonvolatile memory built in the scope. When the scope is connected to the image signal processing unit on the user side, the correction coefficient data is read out to the image signal processing unit side, where the pixel signal obtained from the imaging sensor of the scope is white based on the correction coefficient data. It will be balanced. However, the optical characteristics and the like do not always match between the user-side image signal processing unit and the reference image signal processing unit, and the user-side image signal processing unit uses the white-side image signal processing unit based on the correction coefficient data of each scope. Even if the balance process is performed, the white balance process is not always optimal.
[0010]
Further, the correction coefficient for white balance processing is particularly deeply related to the color temperature characteristics of the light source lamps of the individual image signal processing units. However, the color temperature characteristics of the light source lamps themselves are deteriorated over time. There is also a problem that it changes.
[0011]
In short, in order for proper white balance processing to be performed properly at all times, it is necessary to periodically perform and update correction coefficient settings for white balance processing for individual image signal processing units on the user side.
[0012]
Therefore, another object of the present invention is an electronic endoscope of the type as described above, and it is possible to perform white balance setting as necessary using a histogram extraction circuit used for controlling the aperture opening degree. It is providing the electronic endoscope which became.
[0013]
[Means for Solving the Problems]
An electronic endoscope according to the present invention comprises a scope and an image signal processing unit adapted to be detachably connected to the scope, and includes a light source provided in the image signal processing unit, Light emitted from the light source is guided to the scope to illuminate the front. The electronic endoscope according to the present invention further includes a solid-state imaging unit provided on the distal end side of the scope and a histogram that develops a histogram based on luminance pixel signals for one frame or one field obtained from the solid-state imaging unit. Extraction means and light quantity adjustment means for adjusting the light quantity of light guided from the light source to the scope based on the histogram developed by the histogram extraction means. In the electronic endoscope according to the present invention, the average luminance value is obtained based on the thinned data from all the data of the histogram developed by the histogram extracting means, and the light quantity adjustment by the light quantity adjusting means is performed according to the average luminance value. It is characterized by being.
[0014]
In the preferred embodiment of the present invention, the thinned-out data from all the data of the histogram developed by the histogram extracting means is composed of luminance pixel signals included in every m (integer) luminance levels from the total data.
[0015]
In the preferred embodiment of the present invention, the minimum luminance level and the maximum luminance level of the previous histogram developed by the histogram extraction means are stored, and the minimum luminance is obtained when obtaining the average luminance value based on the next histogram. Decimated data from all data in the extended range including the level and the maximum luminance level is used.
[0016]
Further, in a preferred embodiment of the present invention, the average luminance value and the minimum luminance level and the maximum luminance level at that time are periodically obtained based on all the data of the histogram developed by the histogram extracting means. The light amount adjustment by the light amount adjusting means is performed according to the average luminance value based on all the data of the histogram.
[0017]
Furthermore, in a preferred embodiment of the present invention, the solid-state imaging means outputs a pixel signal of the three primary colors, and the scope stores correction coefficient data for each of the three primary colors specific to itself for white balance processing. A histogram based on the pixel signals for one frame or one field of each of the three primary colors when white balance is set, is developed in the histogram extraction means, and the average value of the corresponding color is obtained from all the histogram data for each color. Then, the correction coefficient data of the other two colors are reset based on the average value of any one of these three primary colors.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
Next, an embodiment of an electronic endoscope according to the present invention will be described with reference to the accompanying drawings.
[0019]
Referring to FIG. 1, an embodiment of an electronic endoscope according to the present invention is shown as a block diagram. The electronic endoscope includes a scope 10 made of a flexible conduit, and the scope 10 is detachably connected to an image signal processing unit 12 called a processor. An imaging sensor 14 composed of a solid-state imaging device, for example, a CCD (charge-coupled device) imaging device, is provided at the distal end or distal end of the scope 10. The imaging sensor 14 has an objective lens system combined with the CCD imaging device. Is included.
[0020]
A light guide 16 comprising an optical fiber bundle is inserted into the scope 10, the distal end of the light guide 16 extends to the distal end of the scope 10, and the end surface of the distal end of the light guide 16 is illuminated. Lenses (not shown) are combined. The proximal end of the light guide 16 is optically connected to a white light source 18 such as a xenon lamp or a halogen lamp provided in the image signal processing unit 12 when the scope 10 is connected to the image signal processing unit 12. A diaphragm 20 and a condenser lens 22 are sequentially provided on the light emission side of the white light source 18, and the diaphragm 20 is used as a light amount adjusting means for appropriately adjusting a light amount from the white light source 18, and the condenser lens 22 is a diaphragm. It is used to collect the light having passed through 20 on the end face of the proximal end of the light guide 16.
[0021]
In the present embodiment, since a frame sequential method is employed to reproduce a color image, a rotary RGB color filter is used as a rotary tri-primary color filter between the proximal end face of the light guide 16 and the condenser lens 22. 24 is interposed. As shown in FIG. 2, the rotary RGB color filter 24 is composed of a disk element, and the disk element is provided with a red filter 24R, a green filter 24G, and a blue filter 24B. It is said. The color filters 24R, 24G, and 24B are arranged along the circumferential direction of the disk element so that the respective radial centers are at an angular interval of 120 °, and an area between adjacent color filters is a light shielding area. Is done.
[0022]
As shown in FIG. 3, the rotary three primary color filter 24 is rotated by a drive motor 26 such as a servo motor or a step motor. The rotation frequency of the rotary RGB color filter 24 is determined according to the TV image reproduction method adopted in the electronic endoscope. For example, when the PAL method is adopted, the rotation frequency of the rotary RGB color filter 24 is 25 Hz, and when the NTSC method is adopted, the rotation frequency is 30 Hz.
[0023]
For example, if the rotary RGB color filter 24 is rotated at a rotation frequency of 30 Hz (NTSC method), the time required for one rotation is approximately 33 ms (1/30 sec), and the illumination time by each color filter is approximately 33/6 ms. Become. From the end face of the distal end of the light guide 16, red light, green light and blue light are emitted in sequence for approximately 33 / 6ms every 33ms (1 / 30sec), and the subject is red light, green light and blue light. Thus, illumination is sequentially performed, and optical subject images of the respective colors are sequentially formed on the light receiving surface of the CCD image sensor by the objective lens system of the image sensor 14. The image sensor 14 photoelectrically converts each color optical subject image formed on the light receiving surface of the CCD image sensor into an analog pixel signal for one frame, and the analog pixel signal for one frame of each color is an illumination time of each color. The readout is sequentially performed from the imaging sensor 14 over the next shading time (33/6 ms) following (33/6 ms), and the readout of the analog pixel signal from the imaging sensor 14 is provided in the scope 10. This is performed by the CCD driver 28.
[0024]
Strictly speaking, since the output power of each color from the color filters 24R, 24G, and 24B and the spectral sensitivity characteristics of the CCD image sensor 14 are different, the illumination times of the red light, the green light, and the blue light are slightly different. Although different, the readout of the analog pixel signal for one frame of each color from the CCD image sensor 14 is performed within the light shielding time in the same manner.
[0025]
As can be seen from FIG. 1, the image signal processing unit 12 is provided with a system controller 30, which is constituted by a microcomputer. That is, the system controller 30 is a central processing unit (CPU), a program for executing various routines, a read only memory (ROM) for storing constants, and a write / readable memory (temporarily storing data). RAM) and controls the overall operation of the electronic endoscope.
[0026]
When the scope 10 is connected to the image signal processing unit 12, the imaging sensor 14 is connected to the CCD process circuit 32 in the image signal processing unit 12 via the CCD driver 28. The analog pixel signal for one frame of each color read from the image sensor 14 by the CCD driver 28 is sent to the CCD process circuit 32, where predetermined image processing such as white balance correction processing, gamma correction processing, contour enhancement processing, etc. is performed. receive. In FIG. 1, the connection relationship between the CCD driver 28 and the CCD process circuit 32 and the system controller 30 is not particularly shown in order to avoid complication, but the pixel signal readout by the CCD driver 28 and the CCD process are not shown. Image processing in the circuit 32 is performed under the control of the system controller 30.
[0027]
The analog pixel signal for one frame of each color processed by the CCD process circuit 32 is sequentially sent to an analog / digital converter 34 where it is converted into a digital pixel signal, and then the digital pixel signal for one frame of each color is converted into a frame memory. 36 is temporarily written and stored. The frame memory 36 is provided with three storage areas for storing digital pixel signals for one frame of each color. From the frame memory 36, digital image signals of three primary colors for one frame are sequentially read out simultaneously. At this time, a horizontal synchronization signal, a vertical synchronization signal, and the like are added to the read digital image signal of each color. That is, digital image signals of three primary colors for one frame are sequentially output from the frame memory 36 as color digital video signals (R, G, B) and sent to the video process circuit 38.
[0028]
The video process circuit 38 is provided with a digital / analog (D / A) and low-pass filter corresponding to the color digital video signal of each color, and the color digital video signal for one frame of each color is color analog video for one frame. The signal is converted into a signal, and then, after passing through a low-pass filter, is appropriately amplified and sent to the color TV monitor device 40, where an optical subject image is reproduced as a color image. The video process circuit 38 generates a component video signal based on the color digital video signals (R, G, B), and the component video signal is a separate TV monitor device, video tape recorder, image processing computer, or the like. Are output from the video process circuit 38 to the outside.
[0029]
Referring to FIG. 4, the diaphragm 20 is illustrated along with its drive mechanism. The diaphragm 20 includes a pair of blade elements 42 and 44, and arm portions 42 </ b> A and 44 </ b> A extend integrally from the blade elements 42 and 44. The blade elements 42 and 44 are pivotally supported by a pivot pin 46 in such a manner that they intersect with each other, and the amount of white light emitted from the white light source 18 is appropriately determined according to the opening degree of the blade elements 42 and 44. Adjusted. The drive mechanism of the diaphragm 20 includes a tension coil spring 48 that is acted between the tips of the arm portions 42A and 44A. The coil spring 48 allows the blade elements 42 and 44 to be elastic so as to reduce the opening thereof. Always receive bias. The pivot pin 46 is appropriately held with respect to the housing of the image signal processing unit 12.
[0030]
The drive mechanism of the diaphragm 20 further includes a cam pin 50 engaged between the arm portions 42A and 44A in order to adjust the opening degree of the pair of blade elements 42 and 44. The cam pin 50 is a lower end portion of the drive plate 52. It is fixed and held on. A rack 54 is formed on one side of the drive plate 52, and a pinion 56 is engaged with the rack 54. The pinion 56 is fixed on an output shaft 58A of an appropriate drive motor 58 such as a servo motor or a step motor. The drive motor 58 is appropriately held with respect to the casing of the image signal processing unit 12, and the rack 54 is appropriately slidably held by appropriate guide means (not shown). When the drive motor 58 is rotated, the cam pin 50 moves up and down together with the drive plate 52, thereby adjusting the opening degree of the blade elements 42 and 44. As shown in FIG. 1, the drive motor 58 is driven by a drive circuit 60, and the drive circuit 60 is controlled by the system controller 30.
[0031]
As shown in FIG. 1, the image signal processing unit 12 is provided with a lamp power circuit 62, and power is supplied to the white light source 18 by the lamp power circuit 62. The lamp power supply circuit 62 is connected to a commercial power supply via a plug (not shown) and is appropriately controlled by the system controller 30.
[0032]
As shown in FIG. 1, the image signal processing unit 12 is provided with a histogram extraction circuit 64. The histogram extraction circuit 64 is one of the frame memory 36 and the video process circuit 38 via a switch (SW) circuit 66. To be selectively connected. The switching of the switch circuit 66 is controlled by the system controller 30, and the histogram extraction circuit 64 is normally connected to the video process circuit 38, so that the luminance pixel signal among the component video signals is output to the histogram extraction circuit 64. It has become. When the histogram extraction circuit 64 is connected to the frame memory 36 by the switch circuit 66, digital image signals of three primary colors for one frame are input to the histogram extraction circuit 64.
[0033]
As shown in FIG. 1, the image signal processing unit 12 is further provided with a character processing circuit 68. The character processing circuit 68 includes a video RAM. Fixed character code information data read from the ROM of the system controller 30 and variable character code information data input from a keyboard (not shown) are temporarily written in predetermined addresses of the video RAM of the character processing circuit 68. The character processing circuit 68 generates a character pattern signal based on the character code information data written in the video RAM, and the character pattern signal is output to the video process circuit and added to the color video signal. Character information is displayed on the device 40 together with a reproduction image of the optical subject image.
[0034]
The character information as described above includes variable character information such as a patient name, examination date and time, medical examination reviews, and fixed character information related to a reproduction color image. Character information particularly related to the present invention includes two fixed character information. For example, there are two messages such as “white balance can be set” and “white balance has been set”. These messages, i.e., fixed character information, are stored in advance in the ROM of the system controller 30 as code data, as described above, and the code data is read from the ROM of the system controller 30 as necessary, and the video RAM of the character processing circuit 68. Is written to. The meaning of the two messages described above will be clarified in the description below.
[0035]
As shown in FIG. 1, a suitable nonvolatile memory such as a rewritable read only memory (EEPROM) 70 is provided on the scope 10 side, and various information of the scope 10 itself is written in the EEPROM 70. For example, the EEPROM 70 stores pixel number data of the CCD image sensor 14, frequency information of a clock pulse when processing an analog image signal read out by the CCD driver 28, and the like, and information particularly related to the present invention. As the data, correction coefficient data G used when white balance processing is performed on the analog pixel signals of the three primary colors. _GAIN , B _GAIN And R _GAIN Is mentioned. When the scope 10 is connected to the image signal processing unit 12, the EEPROM 70 is connected to the system controller 30. At this time, the system controller 30 reads information data in the EEPROM 70 and stores the information data in the RAM in the system controller 30. Retained.
[0036]
The white balance correction process performed by the CCD process circuit 32 is performed by the correction coefficient data G described above. _GAIN , B _GAIN And R _GAIN However, the correction coefficient data G _GAIN , B _GAIN And R _GAIN As mentioned earlier, it is not always optimal.
[0037]
The image signal processing unit 12 is provided with an operation panel 72. The operation panel 72 is provided with various indicator lamps and various switches. In FIG. 1, as switches particularly relevant to the present invention, a power switch (SW) for switching on / off a main power source (not shown) of the image signal processing unit 12 and a lighting switch (SW) for controlling lighting of the light source lamp 18. Reference numerals 74, 76, and 78 denote a mode selector switch (SW) for switching between a normal operation mode and a white balance setting mode of the electronic endoscope and a white balance setting switch (SW) for setting a white balance correction coefficient, respectively. And 80.
[0038]
The switch circuit 66 described above is switched by mode selection by the mode switch 78, and when the normal operation mode is selected by the mode switch 78, the histogram extraction circuit 64 is connected to the video process circuit 38 by the switch circuit 66. When the white balance setting mode is selected by the mode switch 78, the histogram extraction circuit 64 is connected to the frame memory 36 by the switch circuit 66.
[0039]
Referring to FIG. 5, there is illustrated a histogram created by the histogram extraction circuit 64 based on a digital luminance pixel signal for one frame obtained from the video process circuit 38. As shown in FIG. 5, in this embodiment, the histogram extraction circuit 64 distributes the digital luminance pixel signal for one frame into 256 luminance levels. In short, in the histogram shown in FIG. 5, 256 luminance levels are shown along the horizontal axis X, and the vertical axis Y shows the number of digital luminance pixel signals corresponding to each luminance level, that is, the frequency. The luminance level 0 corresponds to the pedestal level, and the luminance level 255 corresponds to the maximum luminance level. In the example shown in FIG. 5, the luminance pixel signal for one frame obtained from the video process circuit 38 includes luminance pixel signals having luminances corresponding to luminance levels 0 to 4 and luminance levels 252 to 255. Absent.
[0040]
Referring to FIGS. 6 and 7, a flowchart of an automatic dimming routine executed by the system controller 30 is shown. This automatic light control routine functions as a subroutine of the main operation routine of the electronic endoscope. For example, when the NTSC system is adopted as the video reproduction system in the electronic endoscope, every 1/30 sec. The interrupt routine to be executed. The execution of the automatic light control routine is started when the power switch 74 on the operation panel 72 is turned on.
[0041]
In step 601, it is determined whether or not the counter N is “0”. Since N = 0 at the initial stage, the process proceeds to step 602, where the average luminance value AI is calculated based on all the data (0 ≦ X ≦ 255) of the first histogram developed by the histogram extraction circuit 64. . That is, the luminance value corresponding to each of the luminance levels 0 to 255 is set to L. ₀ Or L ₂₅₅ And the frequency (number) of luminance pixel signals assigned to each of the luminance levels 0 to 255 is S ₀ Or S ₂₅₅ The average luminance value AI is calculated by the following equation.
[0042]
[Expression 1]

That is, TP is the total number of luminance pixel signals for one frame.
[0043]
In step 603, the drive motor 58 is driven to rotate by the drive circuit 60 in order to give the diaphragm 20 the diaphragm opening corresponding to the calculated average luminance value AI. Next, in step 604, the minimum luminance level L is calculated from the first histogram. _MIN And the maximum luminance level L _MAX Is obtained and stored in the RAM of the system controller 30. FIG. 8 shows an example of the first histogram. As is clear from this example, the minimum luminance level L _MIN There is no luminance pixel signal corresponding to a lower luminance level and the maximum luminance level L _MAX There is no luminance pixel signal corresponding to a luminance level exceeding.
[0044]
In step 605, the count number of the counter N is incremented by “1”, and then in step 606, it is determined whether or not the count number of the counter N is equal to 7. At this stage, since N is 7 or less, that is, N = 1, this routine is temporarily terminated.
[0045]
This routine is executed again after 1/30 sec. At this time, it is assumed that the second histogram as shown in FIG. At this stage, since N = 1, the process proceeds from step 601 to step 607, where the first minimum luminance level L _MIN Is determined as to whether the brightness level is 8 or more. If L _MIN If ≧ 8, proceed to step 608 where the first maximum luminance level L _MAX Is determined to be less than or equal to the luminance level 247. If L _MAX If ≦ 247, the process proceeds to step 609.
[0046]
In step 609, a specific range of the second histogram, that is, the minimum luminance level L of the first histogram. _MIN And maximum brightness level L _MAX The average luminance level AI is determined based on the data thinned out from every other data in the range extended by 8 luminance levels on both sides of the range between the two (the range indicated by the dashed arrow in FIG. 9). Calculated. That is, the average luminance level AI is calculated by the following formula.
[0047]
[Expression 2]

[X] is a Gaussian symbol and represents the maximum integer not exceeding x.
[0048]
In short, the histograms of the luminance pixel signals of two consecutive frames, ie, the first histogram (FIG. 8) and the second histogram (FIG. 9) are generally likely to be similar to each other. (Because there is little movement of the scope when observing the affected area with the scope 10), the previous minimum luminance level L _MIN And maximum brightness level L _MAX Is expanded by 3 to 5% of the histogram definition area (0 ≦ X ≦ 255) on each boundary side (8 luminance levels in this embodiment), the second range is included in the expanded range. It can be assumed that the entity part of the th histogram can be included. Therefore, the calculation range when calculating the average luminance level AI for the second histogram is L _MIN −8 ≦ X ≦ L _MAX Since the calculation is based on thinned data every other time, the calculation time of the average luminance level AI can be greatly shortened.
[0049]
When the average luminance level AI of the second histogram is calculated in step 609, the process proceeds from step 609 to step 603, where the drive motor 58 is driven so as to give the diaphragm 20 the aperture opening corresponding to the average luminance value AI. The circuit 60 is rotationally driven. Next, in step 604, the minimum luminance level L is calculated from the second histogram. _MIN And the maximum luminance level L _MAX Is obtained and stored in the RAM of the system controller 30.
[0050]
In step 605, the count number of the counter N is incremented by “1”, and then in step 606, it is determined whether or not the count number of the counter N is equal to 7. Even at the present stage, since N is 7 or less, that is, N = 2, this routine is temporarily terminated.
[0051]
Step 607 if L _MIN If <8, that is, if the first histogram is as shown in FIG. 10, the process proceeds from step 607 to step 610, where the first maximum luminance level L _MAX Is determined to be less than or equal to the luminance level 247. If L _MAX If ≦ 247, the process proceeds to step 611. For the same reason as described above, if the first histogram is as shown in FIG. 10, it is assumed that the second histogram is similar to that shown in FIG. obtain.
[0052]
In step 611, a specific range of the second histogram shown in FIG. 11, that is, the minimum luminance level L of the first histogram. _MIN And maximum brightness level L _MAX Minimum brightness level L than the range between _MIN Extended to the brightness level 0 and the maximum brightness level L _MAX On the other hand, the average luminance level AI is calculated based on data thinned out from every other data in a range expanded by eight luminance levels (range indicated by broken line arrows in FIG. 11). That is, the average luminance level AI is calculated by the following formula.
[0053]
[Equation 3]

[0054]
As in the case described above, the calculation range when calculating the average luminance level AI for the second histogram is 0 ≦ X ≦ L _MAX Since the calculation is based on thinned data every other time, the calculation time of the average luminance level AI can be greatly shortened.
[0055]
When the average luminance level AI of the second histogram is calculated in step 611, the process proceeds from step 611 to step 603, where the drive motor 58 is driven so as to give the diaphragm 20 the aperture opening corresponding to the average luminance value AI. The circuit 60 is rotationally driven. Next, in step 604, the minimum luminance level L is calculated from the second histogram. _MIN And the maximum luminance level L _MAX Is obtained and stored in the RAM of the system controller 30.
[0056]
In step 605, the count number of the counter N is incremented by “1”, and then in step 606, it is determined whether or not the count number of the counter N is equal to 7. At this stage, since N = 2, this routine is temporarily terminated.
[0057]
Step 608 if L _MAX If> 247, that is, if the first histogram is as shown in FIG. 12, the process proceeds from step 608 to step 612. For the same reason as described above, if the first histogram is as shown in FIG. 12, it is assumed that the second histogram is similar to that, that is, as shown in FIG. obtain.
[0058]
In step 612, a specific range of the second histogram shown in FIG. 13, that is, the minimum luminance level L of the first histogram. _MIN And maximum brightness level L _MAX Minimum brightness level L than the range between _MIN Is expanded by 8 brightness levels on the side and the maximum brightness level L _MAX On the other hand, the average luminance level AI is calculated based on data thinned out from every other data in the range extended to the luminance level 255 (the range indicated by the dashed arrow in FIG. 13). That is, the average luminance level AI is calculated by the following formula.
[0059]
[Expression 4]

[0060]
As in the case described above, the calculation range when calculating the average luminance level AI for the second histogram is L _MIN Since the calculation is narrowed to −8 ≦ X ≦ 255 and the calculation is based on every other thinning data, the calculation time of the average luminance level AI can be greatly shortened.
[0061]
When the average luminance level AI of the second histogram is calculated in step 612, the process proceeds from step 612 to step 603, where the drive motor 58 is driven so as to give the aperture 20 the aperture opening corresponding to the average luminance value AI. The circuit 60 is rotationally driven. Next, in step 604, the minimum luminance level L is calculated from the second histogram. _MIN And the maximum luminance level L _MAX Is obtained and stored in the RAM of the system controller 30.
[0062]
In step 605, the count number of the counter N is incremented by “1”, and then in step 606, it is determined whether or not the count number of the counter N is equal to 7. At this stage, since N = 2, this routine is temporarily terminated.
[0063]
Step 610 If L _MAX If> 247, that is, if the first histogram is as shown in FIG. 14, the process proceeds from step 610 to step 613. For the same reason as described above, if the first histogram is as shown in FIG. 14, it is assumed that the second histogram is similar to that shown in FIG. obtain.
[0064]
In step 613, every other data is thinned out from all the data in the entire domain of the second histogram shown in FIG. 15 (the range indicated by the broken arrow in FIG. 15, ie, 0 ≦ X ≦ 255). Based on this, an average luminance level AI is calculated. That is, the average luminance level AI is calculated by the following formula.
[0065]
[Equation 5]

[0066]
In this case, the calculation range for calculating the average luminance level AI for the second histogram is the entire definition area (0 ≦ X ≦ 255), but the calculation is based on thinned data every other one. Therefore, the calculation time of the average luminance level AI can be greatly shortened.
[0067]
When the average brightness level AI of the second histogram is calculated in step 613, the process proceeds from step 613 to step 603, where the drive motor 58 is driven so as to give the aperture 20 the aperture corresponding to the average brightness value AI. The circuit 60 is rotationally driven. Next, in step 604, the minimum luminance level L is calculated from the second histogram. _MIN And the maximum luminance level L _MAX Is obtained and stored in the RAM of the system controller 30.
[0068]
In step 605, the count number of the counter N is incremented by “1”, and then in step 606, it is determined whether or not the count number of the counter N is equal to 7. At this stage, since N = 2, this routine is temporarily terminated.
[0069]
Subsequently, the average luminance AI of the third to eighth histograms is calculated in any of the above-described manners (steps 609, 611, 612, and 613), and the opening degree of the diaphragm 20 is accordingly changed. Adjust as appropriate. After adjusting the opening degree of the diaphragm 20 based on the average luminance AI obtained from the eighth histogram, the count number of the counter N reaches 7 (step 605). Accordingly, the process proceeds from step 606 to step 614, where the counter N is reset, and this routine is temporarily terminated.
[0070]
This routine is executed again after 1/30 sec. However, since N = 0 at this time, the routine proceeds to step 602 and all the data (0 ≦ 0) of the ninth histogram developed by the histogram extraction circuit 64 is reached. The average luminance value AI is calculated based on X ≦ 255), and the arithmetic expression used at this time is the above-described Expression 1.
[0071]
At this time, the minimum luminance level and the maximum luminance level after reading all the data of the histogram are obtained. Therefore, for example, even when the brightness of the endoscopic image suddenly changes due to the movement of the scope 10 and the range where the image data exists is suddenly different, or when the range where the image data exists is widened or narrowed, Since all data of the histogram (0 ≦ X ≦ 255) is used for calculating the average luminance value AI every eighth time, an appropriate average luminance value can be obtained, and thus the opening of the diaphragm 20 can be adjusted appropriately.
[0072]
Referring to FIGS. 16 to 18, a flowchart of a white balance setting routine executed by the system controller 30 is shown. This white balance setting routine also functions as a subroutine of the main operation routine of the electronic endoscope. For example, when the NTSC system is adopted as the video reproduction system in the electronic endoscope, every 1/30 sec. The interrupt routine to be executed. The execution of the white balance setting routine is performed by selecting the white balance setting mode with the mode switch 78 after the power switch 74 is turned on.
[0073]
Regarding the setting of white balance, for example, when the electronic endoscope according to the present invention is newly delivered to a medical institution such as a hospital together with a plurality of scopes (10), the electronic endoscope according to the present invention is delivered. This is performed when a new scope is added later, or when white balance correction coefficient data is reset for a scope that has already been used.
[0074]
In steps 1601 to 1604, it is sequentially determined whether or not the flags F1, F2, F3, and F4 are “0”. Since F1, F2, F3, and F4 = 0 in the initial stage, the process proceeds to step 1605, where the connection of the switch circuit 66 is switched to the frame memory 36 side. Accordingly, the histogram extraction circuit 64 is connected to the frame memory 36 via the switch circuit 66.
[0075]
In step 1606, it is determined whether or not the light source lamp 18 is turned on by turning on the lighting switch 76. When the light source lamp 18 is not lit, this routine is temporarily terminated. After 1/30 sec, this routine is executed again, but there is no progress unless the light source lamp 18 is turned on.
[0076]
When the lighting of the light source lamp 18 is confirmed in step 1606, the process proceeds to step 1607, where it is determined whether or not a predetermined time has elapsed after the light source lamp 18 is turned on. Execution of this routine is repeated every 1/30 sec, but no progress is made until a predetermined time has elapsed. In other words, the light source lamp 18 is in a standby state until the light emission state is stabilized. When the light emission state of the light source lamp 18 is stabilized after a predetermined time has passed, the process proceeds to step 1608, where a message indicating that white balance can be set, for example, a message such as “white balance can be set” is displayed on the TV. It is displayed on the monitor device 40. The message is displayed by reading out the fixed character information code data stored in advance in the ROM of the system controller 30, writing it into the video RAM of the character processing circuit 68, and outputting it to the video process circuit 38 as a character pattern signal. As mentioned above.
[0077]
In step 1609, it is determined whether or not the scope (10) targeted for white balance setting is connected to the image signal processing unit 12. When the scope is not connected, the routine proceeds to step 1610, where the flag F4 is rewritten from “0” to “1”, and then this routine is once ended. This routine is executed again after 1/30 sec. However, since F4 = 1 at this time, steps 1605 to 1608 are skipped. Thereafter, the execution of this routine is repeated every 1/30 sec. However, there is no progress unless the scope is connected to the image signal processing unit 12.
[0078]
If the connection of the scope is confirmed in step 1609, the process proceeds from step 1609 to step 1611, where the correction coefficient data G is read from the EEPROM 70 of the scope. _GAIN , B _GAIN And R _GAIN Are read and these correction coefficient data are stored in the RAM of the system controller 30. Next, in step 1612, the flag F3 is rewritten from “0” to “1”, and then in step 1613, it is determined whether or not the white balance setting switch 80 is turned on. The white balance setting switch 80 is turned on by manual operation of the operator after completion of white balance setting preparation.
[0079]
Here, the white balance setting preparation will be briefly described. When white balance is set, an enclosure having a reference white color is used. That is, a predetermined reference white color is applied to the inner side wall of the enclosure, and the preparation of white balance is completed by inserting the tip of the scope (10) into the enclosure. Note that white balance setting itself in an electronic endoscope is well known.
[0080]
If it is confirmed in step 1613 that the white balance setting switch 80 is turned on, the process proceeds to step 1614, where the green color developed by the histogram extraction circuit 64 based on the green pixel signal for one frame read from the image sensor 14. The average value AG of the green pixel signal is calculated from all the data of the histogram (G) based on the following formula.
[0081]
[Formula 6]

Where GL _n Indicates 256 green levels for the green pixel signal for one frame, and GS _n Indicates the number of green pixel signals having a value corresponding to each green level, that is, the number of green pixel signals, and TP indicates the total number of green pixel signals for one frame as in the case of Equation 1 described above. The green pixel signal for one frame is the correction coefficient data G obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0082]
In step 1615, the average value AB of the blue pixel signals is calculated from the total data of the blue histogram (B) developed by the histogram extraction circuit 64 based on the blue pixel signal for one frame read from the image sensor 14 as follows. Calculated based on mathematical formula.
[0083]
[Expression 7]

Where BL _n Indicates 256 blue levels for one frame of blue pixel signal, and BS _n Indicates the frequency, that is, the number of blue pixel signals having a value corresponding to each blue level, and TP indicates the total number of blue pixel signals for one frame in the same manner as described above. The blue pixel signal for one frame is the correction coefficient data B obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0084]
In step 1616, a difference ΔGB between the average value AG and the average value AB is calculated. Next, at step 1617, it is determined whether or not the difference ΔGB exceeds a predetermined allowable value PV. When the difference ΔGB exceeds the allowable value PV, the routine proceeds to step 1618, where it is determined whether or not the average value AB is larger than the average value AG. If AB> AG, the process proceeds to step 1619, where correction coefficient data B _GAIN Is decreased by a predetermined value, and this routine is terminated once. On the other hand, if AB <AG, the process proceeds to step 1620, where correction coefficient data B _GAIN Is increased by a predetermined value, and this routine is once terminated.
[0085]
When this routine is executed again in the next interrupt process, after steps 1601 and 1602, the steps 1603 to 1613 are immediately skipped (F3 = 1). At this time, the white balance setting switch 80 is already turned on. Therefore, the process proceeds to step 1614, where the average value AG of the green pixel signal is calculated from the green histogram (G) developed by the histogram extraction circuit 64 again based on the above-described equation (6). At this time, the green histogram is based on the green pixel signal for one frame read next to the green pixel signal for the previous frame read from the image sensor 14, and the latter one frame. Minute green pixel signal is the correction coefficient data G obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0086]
Subsequently, also in step 1615, the average value AB of the blue pixel signal is calculated based on the above-described equation 7 from the blue histogram (B) developed again by the histogram extraction circuit 64. At this time, the blue histogram is based on the blue pixel signal for one frame read next to the blue pixel signal for the previous frame read from the image sensor 14, and the latter one frame. The blue pixel signal of the minute is corrected coefficient data B which has been decreased or increased by a predetermined value in step 1619 or step 1620 at the previous execution of this routine. _GAIN Is white balance processed.
[0087]
In step 1616, the difference ΔGB between the average value AG and the average value AB is calculated again, but the difference ΔGB is smaller than the previous time. This is because the correction coefficient data B in which the white balance processing of the blue pixel signal for one frame this time is decreased or increased by a predetermined value in step 1619 or step 1620 at the previous execution of this routine. _GAIN It is because it is performed based on. Next, in step 1617, it is determined again whether or not the difference ΔGB exceeds a predetermined allowable value PV. When the difference ΔGB still exceeds the allowable value PV, the routine proceeds to step 1618, where it is determined again whether or not the average value AB is larger than the average value AG. If AB> AG at this time, step 1619 is reached. Then, correction coefficient data B _GAIN Is further decreased by a predetermined value, and if AB <AG, the process proceeds to step 1620, where correction coefficient data B _GAIN Is further increased by a predetermined value. In short, correction coefficient data G _GAIN As a reference, the routine described above is repeated until the difference ΔGB falls within the predetermined allowable value PV.
[0088]
When the difference ΔGB falls within the predetermined allowable value PV in step 1617, the process proceeds from step 1617 to step 1621, where the flag F2 is rewritten from “0” to “1”. Next, in step 1622, the average value AG of the green pixel signals is obtained from all the data of the green histogram (G) developed by the histogram extraction circuit 64 based on the green pixel signal for one frame read from the image sensor 14. It is calculated based on the above formula 6. As in the case described above, the green pixel signal for one frame is the correction coefficient data G obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0089]
In step 1623, the average value AR of the red pixel signal is calculated from all the data of the red histogram (R) developed by the histogram extraction circuit 64 based on the red pixel signal for one frame read from the image sensor 14 as follows. Calculated based on mathematical formula.
[0090]
[Equation 8]

Where RL _n Indicates 256 red luminance levels for one frame of red pixel signal, and RS _n Indicates the frequency or number of red pixel signals having a value corresponding to each red level, and TP indicates the total number of red pixel signals for one frame as in the case described above. The red pixel signal for one frame is the correction coefficient data R obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0091]
In step 1624, a difference ΔGR between the average value AG and the average value AR is calculated. Next, at step 1625, it is determined whether or not the difference ΔGR exceeds a predetermined allowable value PV. When the difference ΔGR exceeds the allowable value PV, the routine proceeds to step 1626, where it is determined whether or not the average value AR is larger than the average value AG. If AR> AG, the process proceeds to step 1627, where the correction coefficient data R _GAIN Is decreased by a predetermined value, and this routine is terminated once. On the other hand, if AR <AG, the process proceeds to step 1628 where the correction coefficient data R _GAIN Is increased by a predetermined value, and this routine is once terminated.
[0092]
When this routine is executed again in the next interrupt process, after step 1601, step 1602 to step 1622 are immediately skipped (F2 = 1), and the green histogram (G) developed by the histogram extraction circuit 64 there again. From the above, the average value AG of the green pixel signal is calculated based on Equation 6 described above. At this time, the green histogram is based on the green pixel signal for one frame read next to the green pixel signal for the previous frame read from the image sensor 14, and the latter one frame. Minute green pixel signal is the correction coefficient data G obtained from the EEPROM 70 of the scope. _GAIN Is white balance processed.
[0093]
Subsequently, also in step 1623, the average value AR of the red pixel signal is calculated from the red histogram (R) developed again by the histogram extraction circuit 64 based on the above-described Expression 8. At this time, the red histogram is based on the red pixel signal for one frame read out next to the red pixel signal for one frame read out from the imaging sensor 14, and the latter one frame. Minute red pixel signal is corrected coefficient data R decreased or increased by a predetermined value in step 1627 or step 1628 at the previous execution of this routine. _GAIN Is white balance processed.
[0094]
In step 1624, the difference ΔGR between the average value AG and the average value AR is calculated again, but the difference ΔGR is smaller than the previous time. This is because the correction coefficient data R in which the white balance processing of the red pixel signal for one frame this time is decreased or increased by a predetermined value in step 1627 or step 1628 during the previous execution of this routine. _GAIN It is because it is performed based on. Next, in step 1625, it is determined again whether or not the difference ΔGR exceeds a predetermined allowable value PV. When the difference ΔGR still exceeds the allowable value PV, the routine proceeds to step 1626, where it is determined again whether or not the average value AR is larger than the average value AG. If AR> AG at this time, step 1627 is reached. To the correction coefficient data R _GAIN Is further reduced by a predetermined value, and if AR <AG, the process proceeds to step 1628 where the correction coefficient data R _GAIN Is further increased by a predetermined value. In short, correction coefficient data G _GAIN As a reference, the routine described above is repeated until the difference ΔGR falls within the predetermined allowable value PV.
[0095]
When the difference ΔGR falls within the predetermined allowable value PV in step 1625, the process proceeds from step 1625 to step 1629, where a message indicating that the white balance setting is completed, for example, a message such as “white balance setting is complete”. Is displayed on the TV monitor device 40. The message is displayed by reading out fixed character information code data stored in advance in the ROM of the system controller 30, writing it in the video RAM of the character processing circuit 68, and outputting it to the video process circuit 38 as a character pattern signal. As mentioned above.
[0096]
In step 1630, correction coefficient data G _GAIN , B _GAIN And R _GAIN Is written in the EEPROM 70 of the scope (10). At this time, the correction coefficient data G _GAIN Is not changed, correction coefficient data B _GAIN And R _GAIN Only the EEPROM 70 may be written. Next, in step 1631, it is determined whether or not the scope (10), that is, the scope for which the white balance has been set or reset, has been detached from the image signal processing unit 12. When the scope (10) is not detached, the process proceeds to step 1632 where the flag F1 is rewritten from “0” to “1”. Thereafter, this routine is executed every 1/30 sec. However, there is no progress until the scope (10) is detached. When it is confirmed in step 1631 that the scope (10) has been detached, the process proceeds to step 1633, where the flag F1, the flag F2, and the flag F3 are each rewritten from “1” to “0”, and this routine is once completed.
[0097]
Thereafter, in step 1609, it is monitored whether another scope, that is, a scope to be set or reset, is connected to the image signal processing unit 12, and if such a scope is connected, The routine described in is repeated again. Note that when the mode changeover switch 78 is changed over to select the normal operation mode, this routine ends.
[0098]
In the embodiment described above, the electronic endoscope is assumed to adopt the frame sequential color method, but the electronic endoscope adopting the color simultaneous method, that is, the fine three primary color filter elements are arranged in a mosaic pattern. It should be understood that the same automatic light control and white balance setting can be performed for an electronic endoscope in which the color filter array is applied to the light receiving surface of the solid-state imaging device. Further, in the above description, for the white balance processing, correction coefficient data for the three primary colors is set from the data of each of the three primary colors (R, G, B). In order to obtain data, color difference signal data may be used instead of color data.
[0099]
In the above-described embodiment, the creation of the histogram is based on the pixel signal for one frame. However, the histogram may be created based on the pixel signal for one field.
[0100]
【The invention's effect】
As is apparent from the above description, in the electronic endoscope according to the present invention, when controlling the aperture opening degree based on the histogram developed by the histogram extraction circuit for the luminance pixel signal for one frame, the histogram Since the average brightness value is calculated from the thinned-out data, and the aperture is controlled based on the average brightness value to perform automatic dimming, the responsiveness of the automatic dimming control is improved and the brightness of the reproduced image is optimized. It becomes possible to perform the conversion quickly. In the electronic endoscope according to the present invention, the white balance can be appropriately set by using a histogram extraction circuit used for controlling the aperture opening. In this case, all the histogram data is not thinned out. Since data is used, it is possible to always perform appropriate white balance processing on pixel signals obtained from individual scopes, and to maintain the color reproducibility of endoscopic images by electronic endoscopes with high quality. it can.
[Brief description of the drawings]
FIG. 1 is a schematic block diagram of an electronic endoscope according to the present invention.
FIG. 2 is a front view of the rotary RGB color filter shown in FIG.
FIG. 3 is a side view showing the rotary RGB color filter shown in FIG. 1 together with its drive motor.
4 is a schematic front view showing the diaphragm shown in FIG. 1 together with its drive mechanism. FIG.
FIG. 5 is a graph conceptually showing an example of a histogram developed by the histogram extraction circuit shown in FIG. 1;
6 is a part of a flowchart for explaining an automatic light control routine executed by the system controller of FIG. 1; FIG.
7 is a remaining portion of a flowchart for explaining an automatic light control routine executed by the system controller of FIG. 1; FIG.
8 is a graph conceptually showing an example of a histogram obtained from a luminance pixel signal for one frame, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
9 is a graph showing a histogram related to the histogram of FIG. 8, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
10 is a graph conceptually showing another example of a histogram obtained from a luminance pixel signal for one frame, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
11 is a graph showing a histogram related to the histogram of FIG. 10, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
12 is a graph conceptually showing still another example of a histogram obtained from a luminance pixel signal for one frame, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
13 is a graph showing a histogram related to the histogram of FIG. 12, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
14 is a graph conceptually showing still another example of a histogram obtained from a luminance pixel signal for one frame, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
15 is a graph showing a histogram related to the histogram of FIG. 14, and is an explanatory diagram for explaining the automatic light control routine of FIGS. 6 and 7. FIG.
FIG. 16 is a part of a flowchart for explaining a white balance setting routine executed by the system controller of FIG. 1;
FIG. 17 is another part of the flowchart for explaining the white balance setting routine executed by the system controller of FIG. 1;
FIG. 18 is the remaining part of the flowchart for explaining the white balance setting routine executed by the system controller of FIG. 1;
[Explanation of symbols]
10 Scope
12 Image signal processing unit
14 Imaging sensor
16 Light guide
18 White light source
20 aperture
22 Condensing lens
24 Rotating RGB color filter
28 CCD driver
30 System controller
32 CCD process circuit
34 Analog / digital (A / D) converter
36 frame memory
38 Video process circuit
40 TV monitor device
64 Histogram extraction circuit
66 Switch circuit
70 EEPROM
74 Power switch
76 Light switch
78 Mode selector switch
80 White balance setting switch

Claims (4)

An electronic endoscope comprising a scope and an image signal processing unit adapted to be detachably connected to the scope,
A light source provided in the image signal processing unit is provided, and light emitted from the light source is guided to the scope to illuminate the front thereof, and further provided on the distal end side of the scope. Solid-state imaging means, histogram extraction means for developing a histogram based on luminance pixel signals for one frame or one field obtained from the solid-state imaging means, and from the light source based on the histogram developed by the histogram extraction means In an electronic endoscope comprising a light amount adjusting means for adjusting the amount of light guided to the scope,
An average luminance value is obtained based on thinned data from all the data of the histogram developed by the histogram extracting means, and the light quantity adjustment by the light quantity adjusting means is performed according to the average luminance value ,
The minimum luminance level and the maximum luminance level of the previous histogram developed by the histogram extraction means are stored, and all the extended ranges obtained by extending the minimum luminance level and the maximum luminance level by 3 to 5% of the histogram definition area, respectively. electronic endoscope thinned-out data from the data and wherein Rukoto be utilized in determining the average luminance value based on the next histogram.   2. The electronic endoscope according to claim 1, wherein the thinned-out data from all the data of the histogram developed by the histogram extracting means is composed of luminance pixel signals included in every m (integer) luminance levels. An electronic endoscope characterized by that. 3. The electronic endoscope according to claim 1, wherein an average luminance value and a minimum luminance level and a maximum luminance level at that time are periodically obtained based on all data of the histogram developed by the histogram extracting means. The electronic endoscope is characterized in that the light amount is adjusted by the light amount adjusting means in accordance with an average luminance value based on all the data of the histogram. 4. The electronic endoscope according to claim 1, wherein the solid-state imaging unit outputs pixel signals of three primary colors, and the scope has each of the three primary colors for white balance processing unique to itself. 5. And a histogram based on pixel signals for one frame or one field of each of the three primary colors at the time of white balance setting is developed in the histogram extraction means, and the histogram of each color is stored. An electronic endoscope, wherein an average value of a corresponding color is obtained from all data, and correction coefficient data of the other two colors are reset based on an average value of any one of these three primary colors.

Images