JP3939466B2 - Image processing apparatus, method, and recording medium - Google Patents

Description

【００７１】
また、規格化後の座標値が（ｘ_P ，ｙ_P ）の画素のＢのデータに対し、座標（ｘ_P ，ｙ_P ）をキーにして、ステップ１０２で取り込んだＢの倍率色収差補正データの中から、対応するＢのｙ方向についての倍率色収差量ΔＢｙ（ｘ_P ，ｙ_P ）を検索し、ｙ方向についての歪曲収差補正後の座標値が（ｘ_P ，ｙ_PB')の画素のＢのデータＢ（ｘ_P ，ｙ_PB')の座標を次式に従って変換することを、全ての画素について行う。

【００７２】
更に、規格化後の座標値が（ｘ_P ，ｙ_P ）の画素のＩＲのデータに対し、座標（ｘ_P ，ｙ_P ）をキーにして、ステップ１０２で取り込んだＩＲの倍率色収差補正データの中から、対応するＩＲのｙ方向についての倍率色収差量ΔＩＲｙ（ｘ_P ，ｙ_P ）を検索し、ｙ方向についての歪曲収差補正後の座標値が（ｘ_P ，ｙ_PIR')の画素のＩＲのデータＩＲ（ｘ_P ，ｙ_PIR')の座標を次式に従って変換することを、全ての画素について行う。

【００７３】
上記により、ｙ方向についての歪曲収差補正、及びｙ方向についてのＲ、Ｂ及びＩＲの倍率色収差補正が行われ、Ｒ，Ｇ，Ｂ，ＩＲの各データが表す画像の各画素の位置は各々独立にｙ方向に各々移動されることになる。
【００７４】
次に、画像の各画素のｙ方向についての本来の位置（以下、この位置を座標値（ｘ_P ，ｙ_P0）で表す）を求める。そして、座標値（ｘ_P ，ｙ_P0）の位置におけるＲの値を、歪曲収差補正及び倍率色収差補正を経たデータＲ（ｘ_P ，ｙ_PR）のうち、ｙ方向に沿って座標値（ｘ_P ，ｙ_P0）を挟んで隣り合っている２つの位置におけるＲのデータに基づいて補間演算によって求める。また、座標値（ｘ_P ，ｙ_P0）の位置におけるＧの値を、歪曲収差補正及び倍率色収差補正を経たデータＧ（ｘ_P ，ｙ_PG）のうち、ｙ方向に沿って座標値（ｘ_P ，ｙ_P0）を挟んで隣り合っている２つの位置におけるＧのデータに基づいて補間演算によって求め、座標値（ｘ_P ，ｙ_P0）の位置におけるＢの値を、歪曲収差補正及び倍率色収差補正を経たデータＢ（ｘ_P ，ｙ_PB）のうち、ｙ方向に沿って座標値（ｘ_P ，ｙ_P0）を挟んで隣り合っている２つの位置におけるＢのデータに基づいて補間演算によって求め、座標値（ｘ_P ，ｙ_P0）の位置におけるＩＲの値を、歪曲収差補正及び倍率色収差補正を経たデータＩＲ（ｘ_P ，ｙ_PIR）のうち、ｙ方向に沿って座標値（ｘ_P ，ｙ_P0）を挟んで隣り合っている２つの位置におけるＩＲのデータに基づいて補間演算によって求める。上記処理を画像の全ての画素について行うことにより、ｙ方向についての歪曲収差補正、倍率色収差補正が完了する。
【００７５】
続いて、歪曲収差補正及び倍率色収差補正をｘ方向について行う。すなわち、画像の中心位置を基準として、画像データの座標値が（ｘ_P ，ｙ_P0）の画素に対し、座標（ｘ_P ，ｙ_P0）をキーにして、歪曲収差補正データの中から対応するｘ方向についての歪曲収差量Ｄｘ（ｘ_P ，ｙ_P0）を検索し（なお、座標（ｘ_P ，ｙ_P0）における歪曲収差量がデータとして記憶されていない場合には、前記座標の周囲の位置における歪曲収差量に基づいて、座標（ｘ_P ，ｙ_P0）における歪曲収差量を補間演算によって求める）、座標（ｘ_P ，ｙ_P0）の画素のデータＲ（ｘ_P ，ｙ_P0）、Ｇ（ｘ_P ，ｙ_P0）、Ｂ（ｘ_P ，ｙ_P0）、ＩＲ（ｘ_P ，ｙ_P0）の座標を次式に従って変換することを、全ての画素について行う。
Ｒ（ｘ_PR',ｙ_P0）←Ｒ（ｘ_P ，ｙ_P0） Ｇ（ｘ_PG，ｙ_P0）←Ｇ（ｘ_P ，ｙ_P0）
Ｂ（ｘ_PB',ｙ_P0）←Ｂ（ｘ_P ，ｙ_P0）ＩＲ（ｘ_PIR',ｙ_P0）←ＩＲ（ｘ_P ，ｙ_P0）
但し、ｘ_PR’＝ｘ_PG＝ｘ_PB’＝ｘ_PIR’＝ｘ_P ＋Ｄｘ（ｘ_P ，ｙ_P0）
【００７６】
また、ｘ方向についての歪曲収差補正前の座標値が（ｘ_P ，ｙ_P0）の画素（ｘ方向についての歪曲収差補正後の座標値が（ｘ_PR',ｙ_P0）の画素）のＲのデータに対し、座標（ｘ_P ，ｙ_P0）をキーにして、Ｒの倍率色収差補正データの中から対応するＲのｘ方向についての倍率色収差量ΔＲｘ（ｘ_P ，ｙ_P0）を検索し（なお、座標（ｘ_P ，ｙ_P0）における倍率色収差量がデータとして記憶されていない場合には、前述のように補間演算によって倍率色収差量を演算する）、ｘ方向についての歪曲収差補正後の座標値が（ｘ_PR',ｙ_P0）の画素のＲのデータＲ（ｘ_PR',ｙ_P0）の座標を次式に従って変換することを、全ての画素について行う。

【００７７】
また、ｘ方向についての歪曲収差補正前の座標値が（ｘ_P ，ｙ_P0）の画素のＢのデータに対し、座標（ｘ_P ，ｙ_P0）をキーにして、Ｂの倍率色収差補正データの中から対応するＢのｘ方向についての倍率色収差量ΔＢｘ（ｘ_P ，ｙ_P0）を検索し、ｘ方向についての歪曲収差補正後の座標値が（ｘ_PB',ｙ_P0）の画素のＢのデータＢ（ｘ_PB',ｙ_P0）の座標を次式に従って変換することを、全ての画素について行う。

【００７８】
更に、ｘ方向についての歪曲収差補正前の座標値が（ｘ_P ，ｙ_P0）の画素のＩＲのデータに対し、座標（ｘ_P ，ｙ_P0）をキーにして、ＩＲの倍率色収差補正データの中から対応するＩＲのｘ方向についての倍率色収差量ΔＩＲｘ（ｘ_P ，ｙ_P0）を検索し、ｘ方向についての歪曲収差補正後の座標値が（ｘ_PIR',ｙ_P0）の画素のＩＲのデータＩＲ（ｘ_PIR',ｙ_P0）の座標を次式に従って変換することを、全ての画素について行う。

【００７９】
上記により、ｘ方向についての歪曲収差補正、及びｘ方向についてのＲ、Ｂ及びＩＲの倍率色収差補正が行われ、Ｒ，Ｇ，Ｂ，ＩＲの各データが表す画像の各画素の位置は各々独立にｘ方向に各々移動される。
【００８０】
次に、画像の各画素のｘ方向についての本来の位置（以下、この位置を座標値（ｘ_P0，ｙ_P0）で表す）を求める。そして、座標値（ｘ_P0，ｙ_P0）の位置におけるＲの値を、歪曲収差補正及び倍率色収差補正を経たデータＲ（ｘ_PR，ｙ_P0）のうち、ｘ方向に沿って座標値（ｘ_P0，ｙ_P0）を挟んで隣り合っている２つの位置におけるＲのデータに基づいて補間演算によって求める。また、座標値（ｘ_P0，ｙ_P0）の位置におけるＧの値を、歪曲収差補正及び倍率色収差補正を経たデータＧ（ｘ_PG，ｙ_P0）のうち、ｘ方向に沿って座標値（ｘ_P0，ｙ_P0）を挟んで隣り合っている２つの位置におけるＧのデータに基づいて補間演算によって求め、座標値（ｘ_P0，ｙ_P0）の位置におけるＢの値を、歪曲収差補正及び倍率色収差補正を経たデータＢ（ｘ_PB，ｙ_P0）のうち、ｘ方向に沿って座標値（ｘ_P0，ｙ_P0）を挟んで隣り合っている２つの位置におけるＢのデータに基づいて補間演算によって求め、座標値（ｘ_P0，ｙ_P0）の位置におけるＩＲの値を、歪曲収差補正及び倍率色収差補正を経たデータＩＲ（ｘ_PIR，ｙ_P0）のうち、ｘ方向に沿って座標値（ｘ_P0，ｙ_P0）を挟んで隣り合っている２つの位置におけるＩＲのデータに基づいて補間演算によって求める。
【００８１】
上記処理を画像の全ての画素について行うことにより、ｘ方向についての歪曲収差補正及び倍率色収差補正が行われ、歪曲収差補正処理及び倍率色収差補正処理が完了する。上記の歪曲収差補正及び倍率色収差補正により、結像レンズ２８の歪曲収差及び倍率色収差に起因するＲ，Ｇ，Ｂ，ＩＲの各データが表す画像の画素ずれが補正される。なお、ステップ１０２、１０４は請求項４に記載の補正手段に対応している。
【００８２】
次のステップ１０６では、ＩＲ光として写真フィルムに照射する光に含まれている可視域の光による、ＩＲ光の透過光量の検出値の変化を補正する（以下、波長分離補正という）ための補正パラメータを取り込む。そしてステップ１０８では、取り込んだ補正パラメータに基づき、ＩＲデータに対して波長分離補正を行う。この波長分離補正は、歪曲収差補正及び倍率色収差補正を経たＲ，Ｇ，Ｂ，ＩＲのデータに基づき各画素毎に次式の演算を行い、全ての画素のＩＲデータを次式から求まるデータＩＲ’に置き換えることで実現できる。
【００８３】
【数１】

【００８４】
なお上式において、Ｂ，Ｇ，Ｒ，ＩＲは演算対象の画素のＢ，Ｇ，Ｒ，ＩＲの値、ＩＲ’は波長分離補正後のＩＲの値を表す。また、ａ₀，ａ₁，ａ₂，ａ₃は波長分離補正を行うための補正パラメータであり、各々演算対象の画素のＢ，Ｇ，Ｒ，ＩＲの値に対する係数である。本実施形態では、フィルタ２３ＩＲの特性に基づいて、ＩＲ光として写真フィルムに照射される光のうち可視光の割合及びその波長域を求め、演算結果に基づいて上記補正パラメータの各係数の値を設定し、ＲＯＭ５０に記憶している。
【００８５】
図３には、ＩＲ光として写真フィルムに照射される光の分光特性の一例を示すが（図３に「ＩＲ」と表記して示す特性）、この特性では、前記ＩＲ光として照射される光のうち可視光の割合は数％程度であり、その波長域はＲに相当する波長域にのみ分布している（この場合、ＩＲ光として照射される光に含まれる可視光の透過光量は画像のＲ濃度に応じて変化する）ので、一例としてａ₀＝０，ａ₁＝０，ａ₂＝−０．０３，ａ₃＝１等の値を設定することができる。
【００８６】
上記の波長分離補正を行うことにより、ＩＲ光として照射される光に可視光が含まれていることに起因するＩＲ光の透過光量の検出値の変化が補正される。なお、ステップ１０６、１０８は請求項１に記載の補正手段に対応している。
【００８７】
ところで、写真フィルムを透過する光は、傷や異物等が付いておらずかつ画像が記録されていない部分（所謂素抜け部分）を透過する場合であっても、フィルムベースにおける屈折によって透過光量が減衰するが、フィルムベースにおける光の屈折率（透過光量の減衰度）は透過光の波長によって異なっており（減衰度の波長依存性）、例として図６に示すように、透過光の波長が長くなるに従って透過光の減衰度が小さくなる特性を示すことが一般的である。そして、この波長−減衰度特性はフィルムベースの材質（写真フィルムの種類）によって異なっている。
【００８８】
このため、本実施形態では写真フィルムの波長−減衰度特性を写真フィルムの種類毎に各々測定し、波長−減衰度特性の測定結果に基づいて、写真フィルムによる減衰度の波長依存性に起因するＲ，Ｇ，Ｂ，ＩＲの透過光量の変動を補正するための減衰度補正量を写真フィルムの種類毎に各々決定し、決定した減衰度補正量を写真フィルムの種類を表す情報と対応させて減衰度補正データとしてＲＯＭ５０に予め記憶している。
【００８９】
ステップ１１０では、フィルムスキャナ１２による写真フィルム読取時に検出された、読取対象の写真フィルムの種類を表す情報を取り込む。なお、写真フィルムの種類は、写真フィルムの製造時にバーコード等の形態で写真フィルムの側縁に記録されるＤＸコードを読み取ることで検出することができる。次のステップ１１２では、ステップ１１０で取り込んだ情報に基づき、読取対象の写真フィルムの種類に対応する減衰度補正データをＲＯＭ５０から取り込む。そしてステップ１１４では、ステップ１１２で取り込んだ減衰度補正データに従い、Ｒ，Ｇ，Ｂ，ＩＲの各データに対し、写真フィルムによる減衰度の波長依存性に起因するＲ，Ｇ，Ｂ，ＩＲの透過光量の変動を画素毎に補正する。
【００９０】
これにより、Ｒ，Ｇ，Ｂ，ＩＲの各データの値は、写真フィルムのフィルムベースにおける屈折による透過光の減衰度が波長に拘わらず一定（例えば減衰度＝０）であると仮定したときのＲ，Ｇ，Ｂ，ＩＲの透過光量に対応する値に補正されることになる。上記のステップ１１０乃至１１４は請求項２に記載の補正手段に対応している。
【００９１】
ところで、レンズの焦点距離はレンズを透過する光の波長に応じて変化（焦点距離の波長依存性）するので、単一の画像に対してＲ，Ｇ，Ｂ，ＩＲの各波長域についての読み取りを各々行う場合、結像レンズ２８による画像の結像位置は、エリアＣＣＤ３０の受光面位置に対して透過光の波長域毎に変化し、各波長域についての読み取り（各波長域毎の透過光量の検出）によって得られるＲ，Ｇ，Ｂ，ＩＲの各データが表す画像の鮮鋭度も各々相違することになる。
【００９２】
本実施形態のように各波長域についての読み取りを順次行う場合、上記の鮮鋭度の相違は、例えば各波長域についての読み取りを行う前に、読み取りを行う波長域についてのフィルムスキャナ１２の結像レンズ２８による画像の結像位置をエリアＣＣＤ３０の受光面位置に自動的に一致させるオートフォーカス（ＡＦ）処理を毎回行うことで解決できるが、代わりに画像の読み取りに時間がかかるという新たな問題が生ずる。
【００９３】
本実施形態では、単一の画像に対する読み取りを開始するときに、所定の基準波長域（例えばＧ）についてのみＡＦ処理を行い、各波長域についてＡＦ処理を各々行うことに代えて、次のステップ１１６、１１８でＲ，Ｇ，Ｂ，ＩＲの各データが表す画像の鮮鋭度が一致するように各データを補正する鮮鋭度補正処理を行う。
【００９４】
すなわち、ステップ１１６では、結像レンズ２８の焦点距離の波長依存性に起因するＲ，Ｇ，Ｂ，ＩＲの各データの鮮鋭度のばらつきを補正するための鮮鋭度補正値をＲＯＭ５０から取り込む。この鮮鋭度補正値は、各データが表す画像に対する高周波成分の強調度を表しており（鮮鋭度の低下は空間周波数の分布において高周波成分の減衰として現れる）、各データが表す画像における高周波成分の減衰度合いが高くなるに従って強調度が高くなるように設定されてＲＯＭ５０に予め記憶されている。
【００９５】
詳しくは、Ｒ，Ｇ，Ｂ，ＩＲの各波長域毎の結像レンズ２８の焦点距離から、基準波長域の光についての結像レンズ２８の焦点距離と非基準波長域（例えばＲ，Ｂ，ＩＲ）の光についての結像レンズ２８の焦点距離との偏差を各非基準波長域について各々求め、この焦点距離の偏差に基づいて高周波成分の減衰度合いを推定し、各非基準波長域毎に、前記高周波成分の減衰度合いに応じて鮮鋭度補正値を各々設定している。
【００９６】
次のステップ１１８では、ステップ１１６で取り込んだ鮮鋭度補正値に基づき、非基準波長域のデータに対して鮮鋭度補正を各々実行する。この鮮鋭度補正は、例えば次の（１）式の演算を行うことで実現できる。
ＱＬ＝Ｑ＋β（Ｑ−ＱＵＳ） …（１）
但し、Ｑは補正対象の画像データ（非基準波長域（例えばＲ又はＢ又はＩＲのデータ）、ＱＬは鮮鋭度補正後の画像データ、ＱＵＳは非鮮鋭マスク画像データ、βは高周波成分の強調度（鮮鋭度補正値）である。
【００９７】
非鮮鋭マスク画像データＱＵＳは、補正対象の画像データＱに対し、例えば画像の角部に存在するｎ×ｎ画素（例えばｎ＝５程度の値を用いることができる）の平均値を、前記ｎ×ｎ画素から成る領域の中心に相当する画素の非鮮鋭マスク信号とすることを、処理対象のｎ×ｎ画素の領域を１画素分ずつ移動させながら繰り返すことで得ることができる。これにより、例として図７（Ａ）に実線で示す処理対象の画像データＱのレスポンス特性（空間周波数の分布）に対し、図７（Ａ）に破線で示すように、高周波帯域での応答が画像データＱよりも低くされたレスポンス特性の非鮮鋭マスク画像データＱＵＳが得られる。
【００９８】
（１）式における（Ｑ−ＱＵＳ）は、画像データＱと非鮮鋭マスク画像データＱＵＳとの差であるので、図７（Ｂ）に示すように高周波帯域に応答のピークが生ずるレスポンス特性（すなわち、図７（Ａ）に実線で示すレスポンス特性と破線で示すレスポンス特性の差に相当するレスポンス特性）となる。従って、（１）式によって求まる鮮鋭度補正後の画像データＱＬのレスポンス特性は、図７（Ｃ）に実線で示す画像データＱのレスポンス特性に対し、図７（Ｃ）に破線で示すように、高周波帯域においてのみ応答が高くされた特性になると共に、該特性において高周波帯域における応答高さｈは高周波成分の強調度（鮮鋭度補正値）βの値に依存し、鮮鋭度補正値βの値が高くなるに従って応答高さｈ（高周波成分の強調度合い）も高くなる。
【００９９】
上記の鮮鋭度補正を非基準波長域の各データに対して各々行うことにより、結像レンズ２８の焦点距離の波長依存性に起因するＲ，Ｇ，Ｂ，ＩＲの各データが表す画像の鮮鋭度を補正することができる。このように、ステップ１１６、１１８は請求項６に記載の補正手段に対応している。
【０１００】
なお、上記では結像レンズ２８の焦点距離の波長依存性に起因する鮮鋭度のばらつきを補正していたが、結像レンズ２８の像面湾曲に起因する同一画像内の鮮鋭度のばらつきも併せて補正するようにしてもよい。この補正は、Ｒ，Ｇ，Ｂ，ＩＲの各データが表す画像中の各部における高周波成分の減衰度合いを推定し、高周波成分の減衰を各部における減衰度合いに応じて補正するための鮮鋭度補正値を、２次元のテーブル上の各部の位置に対応するアドレスに各々記憶することにより鮮鋭度補正テーブルを設定し、鮮鋭度の補正にあたり、各画素毎に、鮮鋭度補正テーブルの対応するアドレスに記憶されている鮮鋭度補正値を用いて前述の（１）式の演算を行うことで実現できる。
【０１０１】
次のステップ１２０では上記の各種補正を経たＲ，Ｇ，Ｂの画像データ及びＩＲデータに基づき、Ｒ，Ｇ，Ｂの画像データが表す処理対象の画像の欠陥部を検出する欠陥部検出処理を行う。まず欠陥部検出処理の説明に先立ち、写真フィルムに傷や異物の付いている箇所のＩＲ光による検出の原理について説明する。
【０１０２】
図８（Ａ）に示すように、写真フィルム上の表面に傷や異物が付いていない箇所に光を照射したときの透過光量は、写真フィルムへの入射光量に対し、写真フィルムによる光の吸収に応じた減衰量だけ減衰する。なお、写真フィルムで光の吸収が生ずる波長域はおおよそ可視光域であり、赤外域のＩＲ光については殆ど吸収されないので、前記傷や異物が付いていない箇所にＩＲ光を照射した場合の透過光量は入射光量から僅かに変化するのみである。
【０１０３】
一方、写真フィルム上の傷が付いている箇所に光を照射した場合、照射された光の一部は傷によって屈折するので、前記傷が付いている箇所に光を照射したときの透過光量（前記箇所を直線的に透過する光の光量）は、写真フィルムへの入射光量に対し、前述した写真フィルムによる光の吸収に起因する減衰に、傷による光の屈折に起因する減衰を加えた減衰量だけ減衰する。なお、図８（Ａ）では光の入射側に傷が付いている場合を示しているが、光の射出側に傷が付いている場合も同様である。
【０１０４】
傷による光の屈折はＩＲ光でも生ずるので、前記傷が付いている箇所にＩＲ光を照射した場合のＩＲ光の透過光量は、傷による光の屈折に起因する減衰に応じた減衰量だけ減衰する。なお傷による光の屈折は、例として図８（Ｂ）にも示すように、傷の規模（深さ等）が大きくなるに伴って顕著となる（可視光もＩＲ光も同様）ので、前記傷が付いている箇所にＩＲ光を照射した場合の透過光量は傷の規模が大きくなるに従って小さくなる。従って、ＩＲ光の透過光量の減衰量に基づいて、写真フィルムに付いている傷の規模も検知することができる。
【０１０５】
また、写真フィルム上の塵埃等の異物がついている箇所に光を照射した場合、照射した光は異物によって反射されるので、異物の大きさや種類（光透過率）にも依存するが、前記異物が付いている箇所に光を照射した場合の光の透過光量は前記異物によって大きく減衰する。異物が付いている箇所に光を照射した場合の透過光量の減衰は、前記箇所にＩＲ光を照射した場合も同様である。
【０１０６】
上記のように、写真フィルムにＩＲ光を透過した場合の透過光量は、写真フィルム上の傷又は異物が付いている箇所でのみ変化し、写真フィルムに画像が記録されていたとしても、該画像の透過濃度の変化の影響を受けないので、写真フィルムにＩＲ光を照射して透過光量を検出することで、写真フィルムに付いている傷や異物を検出できる。
【０１０７】
上記に基づき、ステップ１２０では以下のようにして欠陥部検出処理を行う。写真フィルムにＩＲ光を照射したときの透過光量は、前述のように通常は画像上の位置に拘わらず略一定となり、写真フィルムに傷又は異物が付いている箇所でのみ低下する（図９参照）。ＩＲデータは処理対象の画像上の各位置におけるＩＲ光の透過光量を表しているので、処理対象の画像上の傷や異物が付いていない箇所におけるＩＲデータが表すＩＲ光の透過光量（例えば透過光量の最大値）を基準値とする。そして、各画素毎にＩＲ光の透過光量を基準値と比較し、基準値に対する透過光量の変化量（低下量）が所定値（傷や異物が付いていない箇所におけるＩＲ光の透過光量の若干の変動を考慮して定めた値）以上の画素を、修正対象の欠陥部に属する欠陥画素として全て検出する。
【０１０８】
またステップ１２０では、検出した欠陥画素を、欠陥画素相互の位置関係（例えば隣接しているか否か）等に基づいて、同一の欠陥部に属する欠陥画素毎に分類し、各欠陥部に関する情報（例えば各欠陥部に属する欠陥画素を表す情報や各欠陥画素におけるＩＲ光の透過光量の低下量等の情報）をＲＡＭ４８等に記憶する。
【０１０９】
なお、上記の欠陥部検出処理は各種の補正を経たＲ，Ｇ，Ｂ，ＩＲの各データを用いて行われるので、写真フィルムについた傷や異物に相当する欠陥部のみを確実かつ精度良く検出できると共に、欠陥部の範囲等の誤検出も防止することができる。ステップ１２０は請求項７に記載の欠陥部検出手段に対応している。
【０１１０】
次のステップ１２２では、欠陥部検出処理によって検出された修正対象の欠陥部に対し、各欠陥部を修正するための修正値を各々演算する。まず、欠陥部修正の原理について説明する。
【０１１１】
図８（Ｂ）に示すように、写真フィルムの乳剤層はＲ，Ｇ，Ｂの各感光層を含んで構成されており、画像が露光記録され現像等の処理が行われた写真フィルム（ネガフィルム）は、Ｒ感光層にＣのネガ像が形成され、Ｇの感光層にＭのネガ像が形成され、Ｂの感光層にＹのネガ像が形成される。そして写真フィルムを透過した可視光のうち、Ｒ光についてはＲ感光層においてＣのネガ像の透過濃度に応じた減衰量だけ減衰（吸収）され、Ｇ光についてはＧ感光層においてＭのネガ像の透過濃度に応じた減衰量だけ減衰（吸収）され、Ｂ光についてはＢ感光層においてＹのネガ像の透過濃度に応じた減衰量だけ減衰（吸収）される。
【０１１２】
ここで、例として図８（Ｂ）に示すように、乳剤面と反対側のバック面に傷が付いている場合、透過光に対するＲ，Ｇ，Ｂの各感光層における光の吸収の比率は傷が付いていない場合と同じである。すなわち、図８（Ｂ）において、写真フィルムへの入射光量をＩ₀、傷が付いていないときのＲ光、Ｇ光、Ｂ光の透過光量を各々Ｉ_0R，Ｉ_0G，Ｉ_0Bとし、傷が付いたときに傷が付いている箇所を直線的に透過して乳剤層に入射する光量をＩ₁（Ｉ₁＜Ｉ₀：Ｉ₀−Ｉ₁が傷による光の減衰分）、傷が付いているときのＲ光、Ｇ光、Ｂ光の透過光量を各々Ｉ_1R，Ｉ_1G，Ｉ_1Bとすると、以下の（２）式の関係が成り立つ。
Ｉ_0R／Ｉ₀≒Ｉ_1R／Ｉ₁ Ｉ_0G／Ｉ₀≒Ｉ_1G／Ｉ₁ Ｉ_0B／Ｉ₀≒Ｉ_1B／Ｉ₁ …（２）
【０１１３】
従って、バック面に傷が付いている箇所に対応する欠陥部は、傷が付いていない場合と比較して輝度のみが変化し、写真フィルムに記録されている画像の色情報は保存されているので、輝度調整方法を適用し欠陥部領域の輝度を調整することで、画像データが表す画像の欠陥部を修正することができる。
【０１１４】
一方、例として図８（Ｃ）に示すように乳剤面に傷が付いている場合、浅い傷であれば各感光層のうちの一部の感光層が削られることで、透過光に対するＲ，Ｇ，Ｂの各感光層における光の吸収の比率は傷が付いていない場合と変化する。また、各感光層が全て剥ぎ取られているような非常に深い傷であれば、透過光に対する各感光層における光の吸収は生じない。従って、何れの場合も（２）式の関係は成立しない。
【０１１５】
このように、乳剤面に傷が付いている箇所に対応する欠陥部は、傷の深さに拘わらず、傷が付いていない場合と比較して輝度及び色が各々変化しており、写真フィルムに記録されている画像の色情報も失われているので、輝度を調整しても欠陥部を精度良く修正することは困難である。このため、乳剤面に傷が付いている箇所に対応する欠陥部の修正には、欠陥部の周囲の領域の情報から補間によって欠陥部の輝度及び濃度を決定する修正方法（補間方法）が適している。なお、写真フィルムに異物が付いていることに起因して生じた欠陥部についても、異物が付いていない場合と比較して輝度及び色が各々変化するので、上記の欠陥部を修正する場合にも補間方法が適している。
【０１１６】
ステップ１２２では、まず修正対象の各欠陥部に対し、補間方法を適用して修正するか輝度調整方法を適用して修正するかを判定するための所定の特徴量を各々演算する。本実施形態では所定の特徴量の一例として、欠陥部におけるＲ光、Ｇ光、Ｂ光の透過光量の変化の相関を表す特徴量を用いている。
【０１１７】
例えば写真フィルムのバック面に傷が付いている場合、例として図９（Ａ）に示すように、傷が付いている箇所においてＲ光、Ｇ光、Ｂ光の透過光量は略同様の変化を示すので、Ｒ光、Ｇ光、Ｂ光の透過光量の変化の相関は高い。一方、写真フィルムの乳剤面に傷がついている場合、例として図９（Ｂ）に示すように、傷が付いている箇所におけるＲ光、Ｇ光、Ｂ光の透過光量の変化は一定せず、Ｒ光、Ｇ光、Ｂ光の透過光量の変化の相関は低い（写真フィルムに異物が付いている場合も同様）。
【０１１８】
図９は典型的なケースを示したものであり、実際には写真フィルムの両面に傷が付いている等のように、何れの修正方法を適用すべきかが明確でないケースも多々存在しているが、Ｒ，Ｇ，Ｂの画像データに被写体の色に関する情報が残っていれば輝度調整方法を適用することが好ましく、前記情報が残っていない場合には補間方法を適用することが望ましいので、欠陥部におけるＲ光、Ｇ光、Ｂ光の透過光量の変化の相関を表す所定の特徴量（例えばＲ光、Ｇ光、Ｂ光の透過光量の変化の微分値の差を積算した値）に基づいて、適用すべき修正方法を個々の欠陥部毎に適正に判定することができる。
【０１１９】
修正対象の個々の欠陥部について、上述した所定の特徴量を各々演算すると、双方の修正方法の適用範囲の設定値（双方の修正方法の適用範囲の境界を表す閾値）を取り込み、各欠陥部の所定の特徴量を前記設定値と各々比較することにより、個々の欠陥部を単位として、補間方法を適用して修正を行うか輝度調整方法を適用して修正を行うかを各々判定する。
【０１２０】
そして、補間方法を適用して修正を行うと判定した欠陥部に対しては、補間方法を適用して修正値を演算する。すなわち、修正対象の欠陥部の輝度及び色を該欠陥部の周囲の領域の輝度及び色から補間によって新たに演算し、補間演算によって求まる欠陥部内の各画素の値（Ｒ，Ｇ，Ｂ毎の濃度値でもよいし、色相・明度・彩度を表す値でもよい）をＤ１、各画素の元の値をＤ２、修正度合いをαとし、欠陥部内の各画素の修正値Ｄ３を（３）式に従って求める。
Ｄ３＝α・Ｄ１＋（１−α）Ｄ２ …（３）
上記処理を、補間方法を適用して修正を行うと判定した欠陥部に対して各々行って、前記欠陥部に対する修正値を各々求める。
【０１２１】
また、輝度調整方法を適用して修正を行うと判定した欠陥部に対しては、輝度調整方法を適用して修正値を演算する。すなわち、修正対象の欠陥部におけるＩＲ光の透過光量の変化量に基づいて欠陥部の輝度修正量を演算し、輝度修正量に応じて修正した欠陥部内の各画素の輝度値をＬ１、各画素の元の輝度値をＬ２、「欠陥部修正度合い」の現在の設定値をαとし、欠陥部内の各画素の修正値（輝度値）Ｌ３を（４）式に従って求める。
Ｌ３＝α・Ｌ１＋（１−α）Ｌ２ …（４）
上記処理を輝度調整方法を適用して修正を行うと判定した欠陥部に対して各々行って、前記欠陥部に対する修正値を各々求める。なお、上記のステップ１２２は請求項７に記載の修正手段に対応している。
【０１２２】
修正対象の全ての欠陥部に対して修正値が各々演算されるとステップ１２４へ移行し、各欠陥部に対する修正値を、欠陥部の位置を表す情報（例えば各欠陥部を構成する欠陥画素のアドレス）と共にイメージプロセッサ４０に通知し、欠陥部修正値決定処理を終了する。
【０１２３】
イメージプロセッサ４０には、上記の欠陥部修正値決定処理におけるステップ１０２〜１１８の補正を経た画像データが入力され（前記補正をイメージプロセッサ４０が行うようにしてもよい）、入力された画像データに対し、制御部４２で欠陥部修正値決定処理が行われることで制御部４２から通知された修正値に従って欠陥部を修正する（詳しくは欠陥部に属する各欠陥画素の値を、通知された修正画素値に置き換える）欠陥部修正処理を行う。これにより、修正対象の全ての欠陥部が自動的に修正されることになる。このように、イメージプロセッサ４０も請求項７に記載の修正手段に対応している。
【０１２４】
またイメージプロセッサ４０は、欠陥部修正処理を行った画像データに対し、制御部４２におけるセットアップ演算によって決定された処理条件で各種の画像処理を行い、Ｉ／Ｏコントローラ３８及びＩ／Ｆ回路５４を介してプリンタ１６へ出力する。これにより、印画紙６８に露光記録される画像から、修正対象として選択された欠陥部が消去される。
【０１２５】
なお、上記では単一のエリアＣＣＤ３０によってＲ，Ｇ，Ｂ，ＩＲの各波長域についての読み取り（透過光量の検出）を各々行う構成のフィルムスキャナ１２を用いた例を説明したが、これに限定されるものではなく、各波長域についての読み取りを２個以上の光電変換素子で行う構成のフィルムスキャナを用いてもよい。例として図１０には、各波長域についての読み取りを、各々異なるラインセンサ７０Ｂ、７０Ｇ、７０Ｒ、７０ＩＲで行う構成のフィルムスキャナ７２が示されている。
【０１２６】
このフィルムスキャナ７２は、光源２０から射出された光が、光拡散ボックス７４によって写真フィルム２６の幅方向に沿って長いスリット状の光束に整形されて写真フィルム２６に照射され、写真フィルム２６を透過した光が結像レンズ２８を介して各ラインセンサ７０Ｂ、７０Ｇ、７０Ｒ、７０ＩＲに各々入射される構成となっている。各ラインセンサの受光面には、入射光のうち検出すべき波長域の光のみ透過させるフィルタが各々設けられており、写真フィルム２６が一定速度で搬送されている状態で、各ラインセンサによる各波長域についての読み取り（各波長域の光の透過光量の検出）が各々行われる。
【０１２７】
上記のフィルムスキャナ７２のように、各波長域についての読み取りを複数の光電変換素子で行う場合、各光電変換素子の配置位置の相違に起因して、各波長域についての読み取りで各々得られた画像情報に画素ずれが発生する。例えばフィルムスキャナ７２では、隣り合うラインセンサの写真フィルム２６上での読取位置の間隔が、読み取りの１周期当りに写真フィルム２６が搬送される距離のｎ倍（ｎは整数）となるように各ラインセンサが各々配置されていれば、各ラインセンサによる読み取りのタイミングをｎ周期ずつずらせば画素ずれは発生しないが、例えばラインセンサの取付位置の誤差等により、前記読取位置の間隔が前記１周期当りのフィルム搬送距離の非整数倍であった場合には、前述の画素ずれが発生する。
【０１２８】
このような場合には、各ラインセンサの読取位置の所期の位置に対する位置ずれ量を予め測定して記憶しておき、位置ずれ補正を行うようにしてもよい。この位置ずれ補正は、例えば写真フィルムの搬送方向と平行な方向（例えば図５（Ａ）におけるｘ方向）についての歪曲収差補正及び倍率色収差補正において、位置ずれ量の測定値分だけ、位置ずれの方向と逆方向に全画素の位置が移動するように補正することで実現できる。これにより、光電変換素子の配置位置の相違に起因する画素位置のずれを補正することができる。
【０１２９】
なお、上述した補正は請求項５に記載の補正手段に対応しており、フィルムスキャナ７２のラインセンサ７０Ｂ、７０Ｇ、７０Ｒは請求項５に記載の第１の光電変換素子に、ラインセンサ７０ＩＲは請求項５に記載の第２の光電変換素子に対応している。
【０１３０】
また、上記では欠陥部修正方法の一例として補間方法及び輝度調整方法を説明したが、これに限定されるものではなく、ローパスフィルタ等をかけることで欠陥部をぼかす、所謂ぼかし方法を適用してもよい。
【０１３１】
また、上記では写真フィルムを透過した光を光電変換することで画像を読み取る構成を説明したが、これに限定されるものではなく、写真フィルムを反射した光を光電変換することで画像を読み取る構成を採用してもよい。また、画像記録材料は写真フィルムに限定されるものではなく、写真フィルム以外の写真感光材料や普通紙、ＯＨＰシート等を画像記録材料として用いても良いことは言うまでもない。
【０１３２】
更に、上記ではプレスキャン時にＲ，Ｇ，Ｂの読み取りを行い、ファインスキャン時にＲ，Ｇ，Ｂ，ＩＲの読み取りを行う例を説明したが、これに限定されるものではなく、ＩＲ読み取りはプレスキャン時にのみ行ってもよいし、プレスキャン時及びファインスキャン時に各々行ってもよい。
【０１３３】
【発明の効果】
以上説明したように本発明は、画像記録材料の画像記録領域に可視光を照射し画像記録領域を透過又は反射した可視光を検出することで得られた可視画像情報、及び、可視域を一部含む波長域の光を非可視光として画像記録領域に照射し画像記録領域を透過又は反射した非可視光を検出することで得られた非可視画像情報の少なくとも一方に対し、光学特性に起因する各画像情報の差異として、可視域を一部含む波長域の光が非可視光として照射されることに起因する非可視光の検出光量の変化を補正するようにしたので、欠陥部の修正精度を向上させることが可能となる、という優れた効果を有する。
【０１３４】
また本発明は、画像記録材料の画像記録領域に可視光を照射し画像記録領域を透過又は反射した可視光を第１検出手段によって検出することで得られた可視画像情報、及び、画像記録領域に非可視光を照射し画像記録領域を透過又は反射した非可視光を第２検出手段によって検出することで得られた非可視画像情報の少なくとも一方に対し、光学特性に起因する各画像情報の差異として、画像記録材料を透過又は反射する光の画像記録材料による減衰度の波長依存性に起因する、各検出手段による検出光量の差異を補正するようにしたので、欠陥部の修正精度を向上させることが可能となる、という優れた効果を有する。
【図面の簡単な説明】
【図１】 本実施形態に係る画像処理システムの概略構成図である。
【図２】 フィルムスキャナの概略構成を示す斜視図である。
【図３】 写真フィルムに照射する光の分光特性の一例を示す線図である。
【図４】 欠陥部修正値決定処理の内容を示すフローチャートである。
【図５】 歪曲収差補正及び倍率色収差補正を説明するための、（Ａ）は画像に対して設定したｘｙ座標系、（Ｂ）はｘ_P ｙ_P 座標系を示す概念図である。
【図６】 写真フィルムを透過する光の波長と透過光に対する減衰度との関係の一例を示す線図である。
【図７】 （Ａ）は画像データＱ及び非鮮鋭マスク画像データＱＵＳのレスポンス特性の一例、（Ｂ）は画像データ（Ｑ−ＱＵＳ）のレスポンス特性の一例、（Ｃ）は画像データＱ及び鮮鋭度補正後の画像データＱＬのレスポンス特性の一例を各々示す線図である。
【図８】 （Ａ）は写真フィルムの傷及び異物が付いていない箇所、傷が付いている箇所、異物が付いている箇所における光の透過を各々示す概念図、（Ｂ）は写真フィルムのバック面に傷が付いている場合、（Ｃ）は写真フィルムの乳剤面に傷が付いている場合の光の透過を各々示す概念図である。
【図９】 （Ａ）はバック面に傷が付いている場合、（Ｂ）は乳剤面に傷が付いている場合のＲ光、Ｇ光、Ｂ光、ＩＲ光の透過光量の変化の一例を示す線図である。
【図１０】 本発明の他の実施形態に係るフィルムスキャナの概略構成図である。
【符号の説明】
１２ フィルムスキャナ
１４ 画像処理装置
２０ 光源
２３ フィルタユニット
２６ 写真フィルム
２８ 結像レンズ
３０ エリアＣＣＤ
４０ イメージプロセッサ
４２ 制御部
７２ 情報記憶媒体[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus, method, and recording medium, and in particular, visible image information obtained by irradiating an image recording material with visible light and invisible image information obtained by irradiating with invisible light. The present invention relates to an image processing apparatus that performs predetermined image processing using the image processing method, an image processing method applicable to the image processing apparatus, and a recording medium on which a program for causing a computer to function as the image processing apparatus is recorded.
[0002]
[Prior art]
Depending on how the photographic film is handled, the emulsion surface or the back surface (the back side of the emulsion surface) may be scratched, but if a portion corresponding to the image recording area of the photographic film is scratched, the photographic film If the image recorded on the camera is output (recorded on an image recording material such as photographic paper or displayed on a display means such as a display), the scratches on the photographic film may be reduced in density depending on the degree of scratches. In many cases, it is clearly visible on an output image as a defective portion such as a streak or a white streak. In addition, even when foreign matter such as dust adheres to the surface of the photographic film, the foreign matter is clearly visually recognized as a defective portion.
[0003]
In surface exposure type photographic printing equipment that exposes and records images on photographic paper by irradiating light on the photographic film and irradiating the photographic paper with light that has passed through the photographic film, the light source and photographic film are used as countermeasures against scratches on the photographic film. A diffusing plate is placed between the diffusing plate and the photographic film is irradiated with light scattered by the diffusing plate. However, with the above technique, it is difficult to erase a defective portion in an output image (an image recorded by exposure on photographic paper), and the defect is only slightly reduced (not noticeable).
[0004]
As a technique applicable to an image reading apparatus configured to read an image recorded on a photographic film by a reading sensor such as a CCD, Japanese Patent Application Laid-Open No. 11-75039 discloses three wavelengths in the visible light region and a non-visible light region. Obtained by reading in the visible light range based on information obtained by reading each in a non-visible light region, reading each photographic film in a wavelength region of at least four wavelengths including one wavelength (for example, infrared region and ultraviolet region). A technique for correcting the obtained image information is disclosed.
[0005]
The light in the visible range (hereinafter referred to as visible light) changes in the amount of transmitted light according to the image density recorded on the photographic film. The amount of transmitted light changes due to partial refraction or reflection. On the other hand, light in the invisible range (hereinafter referred to as invisible light) is not affected by the image density recorded on the photographic film, although the amount of transmitted light changes where the photographic film has scratches or foreign objects. .
[0006]
Therefore, according to the technique described in the above publication, scratches and foreign matter attached to the photographic film are detected from changes in the amount of transmitted light of invisible light, and visible light transmission caused by the scratches and foreign matter attached to the photographic film is detected. Correcting fluctuations in the amount of light, that is, correcting a defective portion of an image (an image represented by image information obtained by detecting a transmitted light amount of visible light) caused by scratches or foreign matter attached to a photographic film It becomes possible.
[0007]
[Problems to be solved by the invention]
However, the visible light information is acquired by detecting the transmitted light amount of visible light, and the invisible image information is acquired by detecting the transmitted light amount of invisible light. In the case of use for detection / correction of image data, there is a problem that a difference between visible image information and invisible image information occurs due to various optical characteristics.
[0008]
  For example, the refractive index of a film base of a photographic film is generally reduced as the wavelength of light increases. For this reason, when light in the infrared wavelength region is used as invisible light, the wavelength region of invisible light (one of the optical properties) is a short wavelength as a means of improving the detection accuracy of scratches on photographic film. A slight shift to the side can be considered. However, as a part of the wavelength range of invisible light enters the visible range, absorption by the cyan dye component of the emulsion layer of photographic film occurs, and the amount of light transmitted to the photographic film as invisible light is transmitted. However, it slightly changes depending on the R density of the image recorded on the photographic film.
[0009]
  In addition, the light transmitted through the photographic film is transmitted through a portion where no image is recorded (so-called blank portion) without scratches or foreign matter, and the transmitted light amount is reduced due to refraction at the film base. Although it attenuates, the refractive index of light (attenuation of transmitted light amount) in the film base varies depending on the wavelength of transmitted light (the wavelength dependency of attenuation is also one of the optical characteristics), and the wavelength of transmitted light is long. In general, the attenuation of the transmitted light is generally reduced. For this reason, the attenuation of transmitted light due to refraction at the film base of the photographic film will be different between visible light and invisible light, and along with this, the amount of fluctuation in the amount of transmitted visible light irradiated to the photographic film, Variations in the amount of transmitted light of invisible light irradiated on the photographic film are different.
[0013]
  AndMentioned aboveChanges in the amount of transmitted light caused by shifting the wavelength range of invisible lightAnd changes in the amount of transmitted light due to the wavelength dependence of the attenuation of transmitted light due to the film baseHas a problem that, when a defective part is detected by using visible image information and non-visible image information and the defective part is corrected, the correction accuracy for the defective part is lowered.
[0014]
  The present invention has been made in view of the above facts.,Lack ofIt is an object to obtain an image processing apparatus, method, and recording medium capable of improving the accuracy of correcting a recess.
[0015]
[Means for Solving the Problems]
  In order to achieve the above object, an image processing apparatus according to a first aspect of the present invention is directed to a first detection means for irradiating an image recording area of an image recording material with visible light and transmitting or reflecting the visible light through the image recording area. Visible image information obtained by detection, and, Light in the wavelength range including part of the visible rangeThe image recording areaToAnd transmitted or reflected through the image recording area.SaidThe acquisition means for acquiring invisible image information obtained by detecting the invisible light by the second detection means, and at least one of the visible image information and the invisible image information is caused by optical characteristics.As a difference between the pieces of image information, a change in the amount of invisible light detected by the second detection means caused by irradiation of light in a wavelength region partially including the visible region as invisible light.And correction means for correcting.
[0016]
  In the first aspect of the invention, visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and, Light in the wavelength range including part of the visible rangeImage recording areaToThe non-visible image information obtained by detecting the non-visible light transmitted or reflected by the projected image recording area by the second detection unit is acquired by the acquisition unit. The first detection means and the second detection means may be included in the image processing apparatus according to the present invention, or may be included in another apparatus (such as an image reading apparatus). When the first detection unit and the second detection unit are included in different devices, the acquisition of the visible image information and the invisible image information by the acquisition unit can be realized by receiving each image information from the other device. .
  The correction means according to the first aspect of the present invention provides a visible difference as at least one of visible image information and invisible image information as a difference between each piece of image information (visible image information and invisible image information) due to optical characteristics. The change in the detected light amount of the invisible light by the second detecting means due to the irradiation of the light in the wavelength region including a part of the region as invisible light is corrected. This correction is performed by, for example, a spectral light amount distribution of invisible light (or a ratio of light having a visible wavelength included in invisible light), which is an optical characteristic related to the wavelength range of invisible light, or the wavelength range of invisible light. Can be realized by measuring in advance the change in the detected light amount of the invisible light caused by the second detection means caused by including a part of the visible range and correcting the detected light amount of the invisible light.
  As a result, it is possible to obtain visible image information and non-visible image information in which the change in the detected light amount of the non-visible light due to the wavelength range of the non-visible light partially including the visible range is corrected. Then, when detecting the defective portion of the image represented by the visible image information and correcting the detected defective portion, as described in claim 7, by using the visible image information and the invisible image information that have been corrected by the correcting means, Due to the change in the amount of detected invisible light due to the fact that the wavelength range of invisible light partially includes the visible range, the range of the defective portion is erroneously detected when detecting the defective portion of the image, When the correction strength is set inappropriately at the time of correction, it is possible to prevent the correction of the defective portion from being inappropriate, and the correction accuracy of the defective portion can be improved.
[0017]
  An image processing apparatus according to a second aspect of the present invention is obtained by irradiating an image recording region of an image recording material with visible light and detecting visible light transmitted or reflected by the image recording region by a first detection means. Acquisition means for acquiring visible image information and invisible image information obtained by irradiating the image recording area with invisible light and detecting the invisible light transmitted or reflected by the image recording area by the second detection means And at least one of the visible image information and the invisible image information, the wavelength of the attenuation by the image recording material of the light transmitted or reflected by the image recording material as a difference between the image information due to optical characteristics And a correction unit that corrects a difference in the amount of light detected by each of the detection units due to the dependency.
[0018]
  In the invention described in claim 2, visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and the image The invisible image information obtained by irradiating the recording area with invisible light and detecting the invisible light transmitted or reflected through the image recording area by the second detection means is acquired by the acquisition means. Then, the correction means determines the wavelength dependence of the attenuation of the light transmitted through or reflected by the image recording material as a difference between the image information due to optical characteristics with respect to at least one of visible image information and invisible image information. The difference in the amount of light detected by each detection means due to the property is corrected. This correction is an optical characteristic related to the image recording material, such as the wavelength dependence of the attenuation of the transmitted or reflected light by the image recording material, or the difference in the amount of light detected by each detection means due to the wavelength dependence of the attenuation. This can be realized by measuring in advance and correcting the difference in detected light amount.
  Thereby, visible image information and non-visible image information in which the difference in the detected light amount due to the wavelength dependence of the degree of attenuation of the transmitted or reflected light by the image recording material can be obtained. Then, when detecting the defective portion of the image represented by the visible image information and correcting the detected defective portion, as described in claim 7, by using the visible image information and the invisible image information that have been corrected by the correcting means, Due to the difference in the amount of light detected due to the wavelength dependence of the attenuation of the transmitted or reflected light by the image recording material, the range of the defective part is erroneously detected when the defective part of the image is detected, or the correction strength is high when correcting the defective part. By improperly setting, it is possible to prevent improper correction of the defective portion, and it is possible to improve the correction accuracy of the defective portion.
[0019]
When a color image is recorded on the image recording material, visible light transmitted or reflected through the image recording area is detected for each of a plurality of wavelength ranges (preferably three or more different wavelength ranges) within the visible range. By doing so, it is common to obtain information representing the amount of transmitted or reflected light (or transmitted or reflected density) for each of the plurality of wavelength ranges as visible image information. For example, the visible image information may be obtained by sequentially irradiating light in the plurality of wavelength regions to the image recording region, and the first detection unit sequentially detects light transmitted or reflected through the image recording region, or the plurality of wavelength regions. Can be obtained by irradiating the image recording area with visible light including the first light, and detecting the visible light transmitted or reflected through the image recording area by decomposing it into light of the plurality of wavelength ranges. .
[0020]
  However, visible image information for each of a plurality of wavelength ranges obtained as described above has a difference due to optical characteristics related to an optical system such as a lens and a sensor and optical characteristics related to an image recording material. For this reason, the visible image information for each of the plurality of wavelength regions obtained by the first detecting unit detecting the visible light transmitted or reflected through the image recording region for each of the plurality of wavelength regions as the visible image information. Claim each3As described above, it is preferable that the correction unit corrects the difference between the pieces of image information caused by optical characteristics with respect to visible image information and non-visible image information for each of a plurality of wavelength ranges. Thereby, it becomes possible to further improve the correction accuracy of the defective part.
[0021]
  Note that when the first detection means and the second detection means are configured to detect visible light or invisible light that is transmitted or reflected through the image recording region of the image recording material and imaged by the imaging lens. Term4As described in the above, the correcting means forms an image as a difference between each piece of image information (that is, visible image information and invisible image information, or visible image information and invisible image information for each of a plurality of wavelength ranges: the same applies hereinafter). Pixel position shift caused by lateral chromatic aberration or distortion of lensAlsoIt is preferable to correct.
[0022]
For this correction, for example, the lateral chromatic aberration or distortion, which is an optical characteristic related to the imaging lens, or the deviation of the pixel positions of visible image information and invisible image information resulting from the lateral chromatic aberration or distortion is measured in advance. This can be realized by correcting the pixel position deviation. This eliminates pixel position shifts caused by magnification chromatic aberration or distortion of the imaging lens without using an expensive imaging lens that suppresses lateral chromatic aberration and distortion to the invisible range, for example. can do.
[0023]
  Therefore, the claims4According to the invention, due to the displacement of the pixel position of each image information caused by the chromatic aberration of magnification or distortion of the imaging lens (the displacement of the position of each pixel of the image represented by each image information on the actual image), It is possible to prevent the range of the defective portion from being erroneously detected when detecting the defective portion of the image, or the range that is shifted from the actual defective portion when correcting the defective portion.
[0024]
  A configuration in which the first detection means detects visible light by the first photoelectric conversion element and the second detection means detects invisible light by a second photoelectric conversion element different from the first photoelectric conversion element; That is, when the first detection means and the second detection means are configured to detect visible light or invisible light by different photoelectric conversion elements,5As described in the above, the correcting means detects the pixel position shift caused by the difference in the arrangement positions of the first photoelectric conversion element and the second photoelectric conversion element as the difference between the image information.AlsoIt is preferable to correct.
[0025]
For example, this correction can be performed by detecting a difference in the arrangement positions of the first photoelectric conversion element and the second photoelectric conversion element, which are optical characteristics related to the sensor, or a shift in the pixel position of each image information due to the difference in the arrangement position. This can be realized by measuring in advance and using it for correction of pixel position deviation. Thereby, the shift of the pixel position due to the difference in the arrangement position of the first photoelectric conversion element and the second photoelectric conversion element can be eliminated.
[0026]
  Therefore, the claims5According to the invention, due to the displacement of the pixel position due to the difference in the arrangement position of the first photoelectric conversion element and the second photoelectric conversion element, the range of the defective part is erroneously detected when detecting the defective part of the image, It is possible to prevent erroneous correction of a range deviated from the actual defective portion when correcting the defective portion. Claims5In the invention, the first photoelectric conversion element and the second conversion element may be separate or may be integrated.
[0027]
  Further, when the first detection means and the second detection means are configured to detect visible light or invisible light that is transmitted or reflected through the image recording region of the image recording material and imaged by the imaging lens, Term6As described in the above, the correction means determines the difference in the sharpness of the image represented by each image information due to the wavelength dependence of the focal length of the imaging lens as the difference in each image information.AlsoIt is preferable to correct.
[0028]
  This correction is performed in advance by measuring, for example, the difference in the sharpness of the image represented by each image information caused by the focal length for each wavelength, which is an optical characteristic related to the imaging lens, or the wavelength dependence of the focal length, This can be realized by using the sharpness correction. Thereby, the difference in sharpness due to the wavelength dependence of the focal length of the imaging lens can be eliminated. Therefore, the claims6According to the invention, due to the difference in sharpness due to the wavelength dependency of the imaging lens, the range of the defective portion is erroneously detected when detecting the defective portion of the image, or the correction strength is inappropriate when correcting the defective portion. By setting, it is possible to prevent improper correction of a defective portion, for example, an unnecessary edge is generated in an image.
[0032]
  According to an eighth aspect of the present invention, there is provided an image processing method obtained by irradiating an image recording area of an image recording material with visible light and detecting visible light transmitted or reflected by the image recording area by a first detection means. Visible image information and light in a wavelength range including a part of the visible range are applied as invisible light to the image recording area, and the invisible light transmitted or reflected by the image recording area is detected by the second detection unit. A wavelength region partially including the visible region as a difference between the image information resulting from optical characteristics with respect to at least one of the visible image information and the invisible image information. Since the change in the detected light amount of the invisible light by the second detection means caused by the irradiation of the invisible light as the invisible light is corrected, the correction accuracy of the defect portion is improved as in the first aspect of the invention. Is possible That.
[0033]
  Claim9In the image processing method according to the invention described above, the visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means. And invisible image information obtained by irradiating the image recording area with invisible light and detecting the invisible light transmitted or reflected through the image recording area by a second detection means, and obtaining the visible image information And at least one of the invisible image information due to optical characteristicsAs the difference in each image information, the difference in the amount of light detected by each detection means due to the wavelength dependence of the attenuation by the image recording material of the light transmitted or reflected by the image recording materialSo that the claims2As with the invention, it is possible to improve the correction accuracy of the defective portion.
[0034]
  The recording medium according to the invention of claim 10 isComputerVisible image information obtained by irradiating the image recording region of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording region by the first detection means, and, Light in the wavelength range including part of the visible rangeThe image recording areaToAnd transmitted or reflected through the image recording area.SaidAcquire invisible image information obtained by detecting invisible light by the second detecting means.Acquisition means,And the at least one of the visible image information and the invisible image information is caused by optical characteristics.As a difference in each image information, a change in the amount of invisible light detected by the second detection means caused by irradiation of light in a wavelength region partially including the visible region as invisible lightTo correctFunction as correction meansA program for making it happen is recorded.
  Since a program for causing a computer to function as the acquisition unit and the correction unit is recorded on the recording medium according to the tenth aspect, the computer stores the program from the recording medium according to the tenth aspect. By reading and executing the computer, the computer functions as the image processing apparatus according to the invention of claim 1, and it becomes possible to improve the correction accuracy of the defective portion as in the invention of claim 1.
[0035]
  According to an eleventh aspect of the present invention, there is provided a recording medium obtained by irradiating an image recording region of an image recording material with visible light and detecting visible light transmitted or reflected by the image recording region by a first detection means. The obtained visible image information and the invisible image information obtained by irradiating the image recording area with invisible light and detecting the invisible light transmitted or reflected through the image recording area by the second detection means are acquired. The attenuation by the image recording material of light transmitted or reflected by the image recording material as a difference between the image information due to optical characteristics with respect to at least one of the visible image information and the invisible image information. A program for functioning as a correction unit that corrects a difference in the amount of light detected by each detection unit due to the wavelength dependency of the degree is recorded.
  Since a program for causing a computer to function as the acquisition unit and the correction unit is recorded on the recording medium according to the eleventh aspect, the computer stores the program from the recording medium according to the eleventh aspect. By reading and executing the computer, the computer functions as the image processing apparatus according to the invention of claim 2 and, like the invention of claim 2, it is possible to improve the correction accuracy of the defective portion.
[0036]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an example of an embodiment of the present invention will be described in detail with reference to the drawings. In the following, as an example, a case where a defective portion caused by scratches or foreign matters attached to the photographic film is corrected will be described.
[0037]
FIG. 1 shows an image processing system 10 according to the present embodiment. The image processing system 10 includes a film scanner 12, an image processing device 14, and a printer 16 connected in series. The film scanner 12 and the image processing device 14 correspond to the image processing device according to the present invention.
[0038]
The film scanner 12 is an image recorded on a photographic light-sensitive material (hereinafter simply referred to as a photographic film) such as a photographic film (for example, a negative film or a reversal film). Image or positive image) and image data obtained by the reading is output, and as shown in FIG. 2, a light source 20 is provided which comprises a halogen lamp or the like and irradiates the photographic film 26 with light. . The light emitted from the light source includes light having a wavelength in the visible light region and light having a wavelength in the infrared region.
[0039]
On the light emission side of the light source 20, an aperture 21 for adjusting the amount of light applied to the photographic film 26, a filter unit 23, and a light diffusion box 22 that uses the light applied to the photographic film 26 as diffused light are sequentially arranged. ing. The filter unit 23 includes a filter 23C that transmits only light in a wavelength region corresponding to R (R light) among incident light, and a filter 23M that transmits only light (G light) in a wavelength region corresponding to G among incident light. FIG. 2 shows four filters: a filter 23Y that transmits only light in the wavelength region corresponding to B (B light) of incident light, and a filter 23IR that transmits light in the infrared region (IR light) of incident light. It is configured to be fitted into a turret 23A that is rotatable along the direction of arrow A.
[0040]
On the opposite side of the light source 20 across the photographic film 26, an imaging lens 28 and an area CCD 30 are arranged in this order along the optical axis L to form an image of light transmitted through the photographic film 26. The area CCD 30 is a monochrome CCD in which a large number of CCD cells each having sensitivity in the visible light region and the infrared region are arranged in a matrix, and the light receiving surface substantially coincides with the imaging point position of the imaging lens 28. Is arranged. A shutter (not shown) is provided between the area CCD 30 and the imaging lens 28.
[0041]
The area CCD 30 is connected to the scanner control unit 33 via the CCD driver 31. The scanner control unit 33 includes a CPU, a ROM (for example, a ROM capable of rewriting stored contents), a RAM, and an input / output port, which are connected to each other via a bus or the like. The scanner control unit 33 controls the operation of each unit of the film scanner 12. The CCD driver 31 generates a drive signal for driving the area CCD 30 and controls the drive of the area CCD 30.
[0042]
The photographic film 26 is conveyed by a film carrier 24 (see FIG. 1, not shown in FIG. 2), and is positioned at a position (reading position) where the screen center of the image coincides with the optical axis L. The scanner control unit 33 rotates the turret 23A of the filter unit 23 so that all the filters 23 including the filter 23IR are sequentially positioned on the optical axis L while the image is positioned at the reading position. The charge accumulation time of the area CCD 30 corresponding to the predetermined reading condition is set in the CCD driver 31, and the diaphragm 21 is moved to a position corresponding to the predetermined reading condition.
[0043]
As a result, the image recording area on the photographic film 26 is sequentially irradiated with light in the wavelength region (R, G, B, or IR) corresponding to each filter 23, and the light transmitted through the image recording area on the photographic film 26 is the area. It is detected by the CCD 30 (more specifically, photoelectrically converted), and is output from the area CCD 30 as a signal representing the amount of transmitted light. The signal output from the area CCD 30 is converted into digital data representing the amount of transmitted light by the A / D converter 32 and input to the image processing device 14. Thus, the film scanner 12 has functions as the first detection means and the second detection means according to the present invention.
[0044]
  Note that the amount of light transmitted in each wavelength region of R, G, and B varies according to the R, G, and B density of the image recorded in the image recording region (the photographic film 26 has scratches and foreign matter attached thereto). However, the amount of transmitted IR light is not affected by the image density and changes only due to scratches, foreign matter, etc.). Therefore, photoelectric conversion of the transmitted light in each of the R, G, and B wavelength ranges corresponds to reading an image. In the following description, the R, G, B, and IR wavelength ranges input to the image processing apparatus 14 Each of R, G, and B data except IR is referred to as image data. Note that R, G, and B image data are visible image information according to the present invention (more specifically, claims).3And the IR data corresponds to the non-visible image information according to the present invention.
[0045]
In general, a film base of a photographic film has a characteristic that the refractive index decreases as the wavelength of transmitted light increases, and attaches to the photographic film as the wavelength of IR light deviates to the longer wavelength side. The accuracy of detecting a defective portion caused by a scratch or a foreign object is reduced. In the present embodiment, in order to ensure the detection accuracy of the defective portion by IR light, as shown in FIG. 3 as an example, as the filter 23IR, the wavelength region of IR light that passes through the filter 23IR and is irradiated to the photographic film. Uses a filter having spectral characteristics that deviate toward the visible region (short wavelength side), and the light irradiated to the photographic film as IR light contains light in the visible region.
[0046]
On the other hand, the scanner correction unit 36 of the image processing apparatus 14 sequentially performs various correction processes such as dark correction, density conversion, and shading correction on the input image data (and IR data). The output end of the scanner correction unit 36 is connected to the input end of the I / O controller 38, and the image data that has been subjected to each process by the scanner correction unit 36 is input to the I / O controller 38. The input end of the I / O controller 38 is also connected to the data output end of the image processor 40, and image data subjected to image processing (details will be described later) is input from the image processor 40.
[0047]
The input end of the I / O controller 38 is also connected to the control unit 42. The control unit 42 includes an expansion slot (not shown). In this expansion slot, a PC card or an IC card (hereinafter collectively referred to as a digital camera card) that can be loaded into a digital still camera, a CD-ROM, A driver (not shown) for reading / writing data (or a program) to / from an information storage medium such as an MO or CD-R, and a communication control device for communicating with other information processing devices are connected. Image data input from the outside via the expansion slot is input to the I / O controller 38.
[0048]
The output end of the I / O controller 38 is connected to the data input end of the image processor 40 and the control unit 42, and is further connected to the printer 16 via the I / F circuit 54. The I / O controller 38 selectively outputs the input image data to each device connected to the output terminal.
[0049]
In the present embodiment, each image recorded on the photographic film 26 is read twice at different resolutions by the film scanner 12. Note that, in the aspect in which the area sensor (area CCD 30) is used as the reading sensor as in the present embodiment, the switching of the reading resolution (obtaining image data having different resolutions at each reading) is fine even during pre-scanning, for example. The image data is read at the same high resolution as at the time of scanning, and post-processing such as pixel thinning or pixel integration is performed on the obtained image data, or at the time of fine scanning, reading is performed multiple times by the area sensor. Can be realized by moving the area sensor by a distance corresponding to 1 / integer of the pixel interval by an actuator such as a piezo element.
[0050]
In the first reading at a relatively low resolution (pre-scan), even when the image density is very low, the reading conditions determined so as not to saturate accumulated charges in the area CCD 30 (irradiate the photographic film 26). Each image is read with the amount of light for each wavelength region of light R, G, and B and the charge accumulation time of the area CCD 30. In the present embodiment, IR reading is not performed during pre-scanning. Data obtained by this prescan (prescan image data) is input from the I / O controller 38 to the control unit 42.
[0051]
The control unit 42 includes a CPU 46, a RAM 48, a ROM 50 (for example, a ROM capable of rewriting stored contents), and an input / output port 52, which are connected to each other via a bus. The control unit 42 calculates an image feature amount such as image density based on the pre-scan image data input from the I / O controller 38, and the film scanner 12 performs re-processing at a relatively high resolution for each image. Reading conditions for reading (fine scanning) are determined, and the determined reading conditions are output to the film scanner 12.
[0052]
Further, the control unit 42 calculates an image feature amount including extraction of a main image area (for example, an area corresponding to a human face (face area)) in the image based on the pre-scanned image data. Various processing conditions for image processing on image data (fine scan image data) obtained by performing fine scanning are automatically determined by a calculation (setup calculation), and the determined processing conditions are output to the image processor 40.
[0053]
Note that the control unit 42 determines whether or not a defective portion caused by a foreign matter such as a scratch or dust attached to the photographic film 26 is generated in the image represented by the image data based on the IR data input from the film scanner 12. A function for searching whether or not, and a function for setting a parameter for the image processor 40 to perform the defect correction process. A display 43, a keyboard 44, and a mouse (not shown) are connected to the bus of the control unit 42.
[0054]
The control unit 42 generates simulation image data by performing image processing equivalent to the image processing performed by the image processor 40 on the fine scan image data on the pre-scan image data based on the calculated processing conditions of the image processing. To do. Then, the generated simulation image data is converted into a signal for displaying an image on the display 43, and the simulation image is displayed on the display 43 based on the signal. Further, when the displayed simulation image is verified by the operator, such as image quality, and information for instructing correction of processing conditions is input as a verification result via the keyboard 44 or the mouse, based on the input information. Recalculate processing conditions for image processing.
[0055]
On the other hand, image data (fine scan image data) input to the I / O controller 38 by performing fine scan on the image by the film scanner 12 is input from the I / O controller 38 to the image processor 40.
[0056]
The image processor 40 enhances sharpness while suppressing graining, color / density correction processing including gradation conversion and color conversion, pixel density conversion processing, hypertone processing for compressing the gradation of the ultra-low frequency luminance component of the image, and graininess. Each image processing circuit is equipped with various image processing such as hyper-sharpness processing, and the input image data is subjected to various image processing according to the processing conditions determined and notified for each image by the control unit 42. Do. Further, the image processor 40 has a function of performing defect portion correction processing according to the parameters set by the control unit 42.
[0057]
When image data that has been subjected to image processing by the image processor 40 is used for recording an image on photographic paper, the image data that has been subjected to image processing by the image processor 40 is transferred from the I / O controller 38 to an I / F circuit. The image data is output to the printer 16 via 54. When the image data after image processing is output to the outside as an image file, the image data is output from the I / O controller 38 to the control unit 42. Thus, the control unit 42 outputs the image data input from the I / O controller 38 for output to the outside to the outside (the driver, the communication control device, etc.) as an image file through the expansion slot.
[0058]
The printer 16 includes an image memory 58, R, G, and B laser light sources 60, and a laser driver 62 that controls the operation of the laser light sources 60. The recording image data input from the image processing device 14 is temporarily stored in the image memory 58, read out, and used for modulation of R, G, and B laser beams emitted from the laser light source 60. The laser light emitted from the laser light source 60 is scanned on the photographic paper 68 through the polygon mirror 64 and the fθ lens 66, and an image is exposed and recorded on the photographic paper 68. The photographic paper 68 on which the image is recorded by exposure is sent to the processor unit 18 and subjected to color development, bleach-fixing, washing with water, and drying. Thereby, the image exposed and recorded on the photographic paper 68 is visualized.
[0059]
  Next, as an operation of the present embodiment, a defect portion correction value determination process executed by the control unit 42 when fine scan image data is input from the scanner 12 to the image processing device 14 will be described. The defect correction value determination process is claimed in claim8,9 is implemented by executing a defect portion correction value determination program by the CPU 46 of the control unit 42. The defect portion correction value determination program is initially stored in the information storage medium 72 (see FIG. 1) together with a program for causing the CPU 46 to execute other processes. Although the information storage medium 72 is shown as a floppy disk in FIG. 1, it may be constituted by a CD-ROM or a memory card.
[0060]
  When the information reading medium (not shown) connected to the control unit 42 is loaded with the information storage medium 72 and an instruction to transfer (install) the program from the information storage medium 72 to the image processing apparatus 14 is given, the information reading apparatus A defective portion correction value determination program or the like is read from the information storage medium 72 and stored in a ROM 50 that can rewrite the stored contents. When the timing for executing the defective portion correction value determination process arrives, the defective portion correction value determination program is read from the ROM 50, and the program is executed by the CPU 46. Thus, the image processing apparatus 14 is claimed in claim 1., 2It functions as an image processing apparatus described in 1. Thus, the information storage medium 72 storing the defect portion correction value determination program or the like is claimed in claim 10., 11It corresponds to the recording medium described in.
[0061]
Hereinafter, the defect portion correction value determination processing will be described with reference to the flowchart of FIG. In step 100, R, G, B image data and IR data of a single image (processing target image) input to the control unit 42 are taken into the RAM 48 or the like. Step 100 corresponds to the acquisition means of the present invention. The next steps 102 to 118 correspond to the correcting means of the present invention.
[0062]
That is, in step 102, correction data for correcting distortion aberration and lateral chromatic aberration of the imaging lens 28 of the film scanner 12 is fetched from the ROM 50. The distortion aberration correction data according to the present embodiment is data used for distortion aberration correction for correcting the geometric distortion of the image caused by the distortion aberration of the imaging lens 28, and the image caused by the distortion aberration of the imaging lens 28. It is set based on the result of measuring the change direction and the change amount of the pixel position at each of the above positions.
[0063]
In the present embodiment, G is adopted as the reference wavelength region, and the amount of distortion (amount of distortion) of the pixel position for G at each position on the image due to the distortion of the imaging lens 28 is measured as distortion correction data. The result is decomposed in the x and y directions, and the amount of distortion at each position on the image is expressed as x_Py_PCoordinate system (center position of image (x_P0, Y_P0) As the origin (= (0,0)) and any pixel on the image is represented as a coordinate value (x_P, Y_P) (Refer to FIG. 5B)) as a reference, the distortion aberration amount Dx (x_P, Y_P) And a distortion aberration amount Dy (x_P, Y_P) Is used.
[0064]
The lateral chromatic aberration correction data is data used for lateral chromatic aberration correction that corrects color misregistration of an image caused by lateral chromatic aberration of the imaging lens 28. At each position on the image caused by lateral chromatic aberration of the imaging lens 28. Is set based on the result of measuring the change direction and change amount of the pixel position of the image represented by the non-reference wavelength region data with respect to the pixel position of the image represented by the reference wavelength region data.
[0065]
In the present embodiment, R, B, and IR are employed as the non-reference wavelength region, and R pixel positions relative to G at each position on the image due to the magnification chromatic aberration of the imaging lens 28 are used as R magnification chromatic aberration correction data. The measurement result of the amount of change (the amount of chromatic aberration of magnification) is decomposed in the x and y directions, and the amount of chromatic aberration of magnification of R at each position on the image is_Py_PAmount of chromatic aberration of magnification ΔRx (x in the x direction of R with reference to the coordinate system_P, Y_P) And R magnification chromatic aberration amount ΔRy (x in the y direction)_P, Y_P) Is used.
[0066]
Further, as the B magnification chromatic aberration correction data, the measurement results of the amount of change in the pixel position of B with respect to G (magnification chromatic aberration amount) at each position on the image due to the magnification chromatic aberration of the imaging lens 28 are expressed in the x and y directions. And the amount of lateral chromatic aberration of B at each position on the image is expressed as x_Py_PWith reference to the coordinate system, the amount of lateral chromatic aberration ΔBx (x_P, Y_P) And B in the y-direction magnification chromatic aberration amount ΔBy (x_P, Y_P) Is used.
[0067]
Furthermore, as IR magnification chromatic aberration correction data, measurement results of the amount of change in the IR pixel position with respect to G (amount of chromatic aberration of magnification) at each position on the image caused by the magnification chromatic aberration of the imaging lens 28 are measured in the x and y directions. The amount of chromatic aberration of magnification of IR at each position on the image is expressed as x_Py_PWith reference to the coordinate system, the amount of lateral chromatic aberration ΔIRx (x_P, Y_P) And IR chromatic aberration amount ΔIRy (x_P, Y_P) Is used.
[0068]
In the next step 104, based on the distortion aberration correction data and magnification chromatic aberration correction data fetched from the ROM 50, distortion aberration / magnification chromatic aberration correction is performed for each of the R, G, B, and IR data in the y and x directions. Do.
[0069]
That is, first, the coordinate value (x, y) of each pixel of the image data is set to x based on the center position of the image._Py_PCoordinate values (x in the coordinate system (see FIG. 5B))_P, Y_P) (X_P= Xx_P0, Y_P= Y-y_P0: After normalization, the coordinate value after normalization is (x_P, Y_P) Pixel coordinates (x_P, Y_P) As a key, the distortion aberration amount Dy (x) in the corresponding y direction from the distortion correction data acquired in step 102._P, Y_P) And coordinates (x_P, Y_P) Pixel data R (x_P, Y_P), G (x_P, Y_P), B (x_P, Y_P), IR (x_P, Y_P) Is converted for all pixels.
R (x_P, Y_PR') ← R (x_P, Y_P) G (x_P, Y_PG) ← G (x_P, Y_P)
B (x_P, Y_PB') ← B (x_P, Y_P) IR (x_P, Y_PIR') ← IR (x_P, Y_P)
However, y_PR'= Y_PG= Y_PB'= Y_PIR'= Y_P+ Dy (x_P, Y_P)
[0070]
Also, the normalized coordinate value is (x_P, Y_P) Pixel (the coordinate value after distortion correction in the y direction is (x_P, Y_PRThe coordinates (x_P, Y_P) As a key, the magnification chromatic aberration amount ΔRy (x in the y direction of the corresponding R from the magnification chromatic aberration correction data of R captured in step 102._P, Y_P) And the coordinate value after distortion correction in the y direction is (x_P, Y_PR') Pixel R data R (x_P, Y_PRThe coordinates of ') are converted according to the following formula for all pixels.

[0071]
Also, the normalized coordinate value is (x_P, Y_P) For pixel B data, the coordinates (x_P, Y_P) As a key, the magnification chromatic aberration amount ΔBy (x in the y direction of the corresponding B from the B magnification chromatic aberration correction data captured in step 102._P, Y_P) And the coordinate value after distortion correction in the y direction is (x_P, Y_PB') Pixel B data B (x_P, Y_PBThe coordinates of ') are converted according to the following formula for all pixels.

[0072]
Furthermore, the normalized coordinate value is (x_P, Y_P) For the IR data of pixel (x)_P, Y_P) As a key, the amount of chromatic aberration of magnification ΔIRy (x_P, Y_P) And the coordinate value after distortion correction in the y direction is (x_P, Y_PIR') IR data IR (x_P, Y_PIRThe coordinates of ') are converted according to the following formula for all pixels.

[0073]
Thus, distortion aberration correction in the y direction and R, B, and IR magnification chromatic aberration correction in the y direction are performed, and the position of each pixel of the image represented by each data of R, G, B, and IR is independent of each other. Respectively moved in the y direction.
[0074]
Next, the original position of each pixel of the image in the y direction (hereinafter, this position is expressed as a coordinate value (x_P, Y_P0)). And the coordinate value (x_P, Y_P0) At the position of R), the data R (x_P, Y_PR), The coordinate value (x_P, Y_P0) Between two positions adjacent to each other with the interpolation calculation. Also, the coordinate value (x_P, Y_P0) At the position G), the data G (x_P, Y_PG), The coordinate value (x_P, Y_P0) Is obtained by interpolation based on G data at two positions adjacent to each other, and a coordinate value (x_P, Y_P0) At the position B), the data B (x_P, Y_PB), The coordinate value (x_P, Y_P0) Between two positions adjacent to each other with the interpolation calculation based on the B data, and the coordinate value (x_P, Y_P0), The IR value at the position of the data IR (x_P, Y_PIR), The coordinate value (x_P, Y_P0) Between two positions adjacent to each other on the basis of IR data. By performing the above processing for all the pixels of the image, the distortion aberration correction and the magnification chromatic aberration correction in the y direction are completed.
[0075]
Subsequently, distortion correction and lateral chromatic aberration correction are performed in the x direction. That is, the coordinate value of the image data is (x_P, Y_P0) Pixel coordinates (x_P, Y_P0) As a key, the distortion aberration amount Dx (x_P, Y_P0) (Note that the coordinates (x_P, Y_P0) Is not stored as data, the coordinates (x) are based on the distortion aberration amounts at positions around the coordinates._P, Y_P0) To determine the amount of distortion in interpolation), coordinates (x_P, Y_P0) Pixel data R (x_P, Y_P0), G (x_P, Y_P0), B (x_P, Y_P0), IR (x_P, Y_P0) Is converted for all pixels.
R (x_PR', y_P0) ← R (x_P, Y_P0) G (x_PG, Y_P0) ← G (x_P, Y_P0)
B (x_PB', y_P0) ← B (x_P, Y_P0) IR (x_PIR', y_P0) ← IR (x_P, Y_P0)
Where x_PR'= X_PG= X_PB'= X_PIR'= X_P+ Dx (x_P, Y_P0)
[0076]
The coordinate value before distortion correction in the x direction is (x_P, Y_P0) Pixel (the coordinate value after distortion correction in the x direction is (x_PR', y_P0) Pixel)) for the R data, the coordinates (x_P, Y_P0) As a key, the amount of chromatic aberration of magnification ΔRx (x_P, Y_P0) (Note that the coordinates (x_P, Y_P0) Is not stored as data, the magnification chromatic aberration amount is calculated by interpolation as described above), and the coordinate value after distortion aberration correction in the x direction is (x_PR', y_P0) Pixel R data R (x)_PR', y_P0) Is converted for all pixels.

[0077]
The coordinate value before distortion correction in the x direction is (x_P, Y_P0) For pixel B data, the coordinates (x_P, Y_P0) As a key, the amount of chromatic aberration of magnification ΔBx (x in the x direction of B corresponding from the magnification chromatic aberration correction data of B_P, Y_P0) And the coordinate value after distortion correction in the x direction is (x_PB', y_P0) Pixel B data B (x)_PB', y_P0) Is converted for all pixels.

[0078]
Further, the coordinate value before distortion correction in the x direction is (x_P, Y_P0) For the IR data of pixel (x)_P, Y_P0) As a key, the amount of chromatic aberration of magnification ΔIRx (x_P, Y_P0) And the coordinate value after distortion correction in the x direction is (x_PIR', y_P0) Pixel IR data IR (x_PIR', y_P0) Is converted for all pixels.

[0079]
As described above, distortion aberration correction in the x direction and R, B, and IR magnification chromatic aberration correction in the x direction are performed, and the positions of the pixels of the image represented by the R, G, B, and IR data are independent of each other. Respectively moved in the x direction.
[0080]
Next, the original position in the x direction of each pixel of the image (hereinafter, this position is expressed as a coordinate value (x_P0, Y_P0)). And the coordinate value (x_P0, Y_P0) At the position of R), the data R (x_PR, Y_P0), The coordinate value (x_P0, Y_P0) Between two positions adjacent to each other with the interpolation calculation. Also, the coordinate value (x_P0, Y_P0) At the position G), the data G (x_PG, Y_P0), The coordinate value (x_P0, Y_P0) Is obtained by interpolation based on G data at two positions adjacent to each other, and a coordinate value (x_P0, Y_P0) At the position B), the data B (x_PB, Y_P0), The coordinate value (x_P0, Y_P0) Between two positions adjacent to each other with the interpolation calculation based on the B data, and the coordinate value (x_P0, Y_P0), The IR value at the position of the data IR (x_PIR, Y_P0), The coordinate value (x_P0, Y_P0) Between two positions adjacent to each other on the basis of IR data.
[0081]
  By performing the above processing for all the pixels of the image, distortion aberration correction and magnification chromatic aberration correction in the x direction are performed, and the distortion aberration correction processing and magnification chromatic aberration correction processing are completed. By the above distortion aberration correction and magnification chromatic aberration correction, the pixel shift of the image represented by the R, G, B, and IR data caused by the distortion aberration and magnification chromatic aberration of the imaging lens 28 is corrected. Steps 102 and 104 are claimed.4This corresponds to the correction means described in (1).
[0082]
In the next step 106, correction for correcting a change in the detected value of the transmitted light amount of IR light due to light in the visible range included in the light irradiated to the photographic film as IR light (hereinafter referred to as wavelength separation correction). Import parameters. In step 108, wavelength separation correction is performed on the IR data based on the acquired correction parameters. In this wavelength separation correction, the following equation is calculated for each pixel based on R, G, B, and IR data subjected to distortion aberration correction and magnification chromatic aberration correction, and IR data of all pixels is obtained from the following equation: It can be realized by replacing with '.
[0083]
[Expression 1]

[0084]
In the above equation, B, G, R, and IR represent the B, G, R, and IR values of the pixel to be calculated, and IR ′ represents the IR value after wavelength separation correction. A₀, A₁, A₂, A_ThreeIs a correction parameter for performing wavelength separation correction, and is a coefficient for the B, G, R, and IR values of each pixel to be calculated. In the present embodiment, based on the characteristics of the filter 23IR, the proportion of visible light and the wavelength range of the light irradiated to the photographic film as IR light are obtained, and the value of each coefficient of the correction parameter is calculated based on the calculation result. It is set and stored in the ROM 50.
[0085]
FIG. 3 shows an example of the spectral characteristics of light irradiated to the photographic film as IR light (characterized as “IR” in FIG. 3). In this characteristic, the light irradiated as the IR light is shown. The ratio of visible light is about several percent, and the wavelength range is distributed only in the wavelength range corresponding to R (in this case, the transmitted light amount of visible light included in the light irradiated as IR light is an image). As an example, a₀= 0, a₁= 0, a₂= -0.03, a_ThreeA value such as = 1 can be set.
[0086]
  By performing the wavelength separation correction described above, the change in the detected value of the transmitted light amount of the IR light due to the fact that visible light is included in the light irradiated as the IR light is corrected. Steps 106 and 108 are claimed.1This corresponds to the correction means described in (1).
[0087]
By the way, the light transmitted through the photographic film has a transmitted light amount due to refraction at the film base even when it is transmitted through a portion where no image is recorded (so-called blank portion) without scratches or foreign matters. Although the light is attenuated, the refractive index of light (attenuation of transmitted light amount) in the film base varies depending on the wavelength of transmitted light (wavelength dependence of attenuation). As an example, as shown in FIG. In general, the attenuation characteristic of transmitted light decreases as the length increases. The wavelength-attenuation characteristics vary depending on the film base material (type of photographic film).
[0088]
For this reason, in this embodiment, the wavelength-attenuation characteristic of the photographic film is measured for each type of photographic film, and based on the measurement result of the wavelength-attenuation characteristic, this results from the wavelength dependence of the attenuation by the photographic film. Attenuation correction amounts for correcting fluctuations in the amount of transmitted light of R, G, B, and IR are determined for each type of photographic film, and the determined attenuation correction amount is associated with information representing the type of photographic film. It is stored in advance in the ROM 50 as attenuation correction data.
[0089]
In step 110, information representing the type of photographic film to be read, which is detected when the film scanner 12 reads the photographic film, is fetched. The type of photographic film can be detected by reading a DX code recorded on the side edge of the photographic film in the form of a barcode or the like when the photographic film is manufactured. In the next step 112, attenuation correction data corresponding to the type of the photographic film to be read is fetched from the ROM 50 based on the information fetched in step 110. In step 114, the R, G, B, IR transmission due to the wavelength dependence of the attenuation by the photographic film is performed on the R, G, B, IR data according to the attenuation correction data fetched in step 112. The variation in the amount of light is corrected for each pixel.
[0090]
  Thus, the values of R, G, B, and IR data are based on the assumption that the attenuation of transmitted light due to refraction at the film base of a photographic film is constant regardless of the wavelength (eg, attenuation = 0). It is corrected to a value corresponding to the transmitted light amount of R, G, B, IR. Said steps 110 to 114 are claimed2This corresponds to the correction means described in (1).
[0091]
By the way, since the focal length of the lens changes according to the wavelength of light transmitted through the lens (the wavelength dependency of the focal length), reading for each wavelength region of R, G, B, and IR with respect to a single image. In this case, the image forming position of the image by the imaging lens 28 changes for each wavelength range of the transmitted light with respect to the light receiving surface position of the area CCD 30, and reading for each wavelength range (transmitted light amount for each wavelength range). The sharpness of the image represented by each of the R, G, B, and IR data obtained by (detection of the above) is also different.
[0092]
When reading is sequentially performed for each wavelength region as in the present embodiment, the difference in the sharpness is, for example, the image formation of the film scanner 12 for the wavelength region to be read before performing the reading for each wavelength region. This can be solved by performing autofocus (AF) processing for automatically matching the image formation position of the image by the lens 28 with the light receiving surface position of the area CCD 30, but a new problem that it takes time to read the image instead. Arise.
[0093]
In the present embodiment, when reading a single image is started, instead of performing AF processing only for a predetermined reference wavelength range (for example, G) and performing AF processing for each wavelength range, the following steps are performed. At 116 and 118, sharpness correction processing is performed to correct each data so that the sharpness of the image represented by each of the R, G, B, and IR data matches.
[0094]
That is, in step 116, a sharpness correction value for correcting the variation in sharpness of each R, G, B, and IR data caused by the wavelength dependence of the focal length of the imaging lens 28 is fetched from the ROM 50. This sharpness correction value represents the enhancement degree of the high frequency component with respect to the image represented by each data (a reduction in sharpness appears as attenuation of the high frequency component in the spatial frequency distribution), and the high frequency component in the image represented by each data. The degree of enhancement is set so as to increase as the degree of attenuation increases, and is stored in advance in the ROM 50.
[0095]
Specifically, from the focal length of the imaging lens 28 for each wavelength region of R, G, B, and IR, the focal length of the imaging lens 28 and the non-reference wavelength region (for example, R, B, IR) with respect to the focal length of the imaging lens 28 is obtained for each non-reference wavelength region, the degree of attenuation of the high-frequency component is estimated based on this focal length deviation, and for each non-reference wavelength region. Each sharpness correction value is set in accordance with the attenuation degree of the high-frequency component.
[0096]
In the next step 118, the sharpness correction is performed on the data in the non-reference wavelength region based on the sharpness correction value fetched in step 116. This sharpness correction can be realized, for example, by performing the following equation (1).
QL = Q + β (Q−QUS) (1)
However, Q is image data to be corrected (non-reference wavelength range (for example, R, B, or IR data), QL is image data after sharpness correction, QUS is unsharp mask image data, and β is an enhancement degree of high frequency components. (Sharpness correction value).
[0097]
The non-sharp mask image data QUS is, for example, an average value of n × n pixels (for example, a value of about n = 5 can be used) existing in the corners of the image with respect to the image data Q to be corrected. It is possible to obtain an unsharp mask signal for a pixel corresponding to the center of a region composed of × n pixels by repeating the region of the n × n pixel to be processed by moving one pixel at a time. Accordingly, as an example, the response characteristic (spatial frequency distribution) of the image data Q to be processed indicated by the solid line in FIG. 7A has a response in the high frequency band as indicated by the broken line in FIG. The non-sharp mask image data QUS having a response characteristic lower than that of the image data Q is obtained.
[0098]
Since (Q−QUS) in the equation (1) is the difference between the image data Q and the unsharp mask image data QUS, as shown in FIG. FIG. 7A shows a response characteristic corresponding to the difference between the response characteristic indicated by the solid line and the response characteristic indicated by the broken line. Accordingly, the response characteristic of the image data QL after sharpness correction obtained by the equation (1) is as shown by a broken line in FIG. 7C, compared to the response characteristic of the image data Q indicated by a solid line in FIG. 7C. In this characteristic, the response height h in the high frequency band depends on the enhancement degree (sharpness correction value) β of the high frequency component, and the sharpness correction value β As the value increases, the response height h (high frequency component enhancement degree) also increases.
[0099]
  By performing the above-described sharpness correction on each data in the non-reference wavelength range, the sharpness of the image represented by the R, G, B, and IR data resulting from the wavelength dependence of the focal length of the imaging lens 28 is obtained. The degree can be corrected. Thus, steps 116 and 118 are claimed.6This corresponds to the correction means described in (1).
[0100]
In the above description, the variation in sharpness due to the wavelength dependence of the focal length of the imaging lens 28 is corrected. However, the variation in sharpness in the same image due to the curvature of field of the imaging lens 28 is also combined. May be corrected. This correction estimates the degree of attenuation of the high-frequency component in each part in the image represented by each data of R, G, B, and IR, and corrects the sharpness correction value for correcting the attenuation of the high-frequency component according to the degree of attenuation in each part. Is stored at addresses corresponding to the positions of the respective parts on the two-dimensional table, and a sharpness correction table is set. For sharpness correction, each pixel is stored at a corresponding address in the sharpness correction table. This can be realized by performing the calculation of the above-described equation (1) using the sharpness correction value.
[0101]
In the next step 120, a defect portion detection process for detecting a defect portion of the image to be processed represented by the R, G, B image data based on the R, G, B image data and IR data subjected to the above various corrections. Do. First, prior to the description of the defect detection process, the principle of detection by IR light of a spot where a scratch or a foreign object is attached to the photographic film will be described.
[0102]
As shown in FIG. 8 (A), the amount of transmitted light when light is applied to a spot on the surface of the photographic film where there are no scratches or foreign matter is absorbed by the photographic film relative to the amount of incident light on the photographic film. Attenuation is performed according to the amount of attenuation. Note that the wavelength range where light absorption occurs in a photographic film is approximately the visible light range, and IR light in the infrared region is hardly absorbed. Therefore, transmission when IR light is applied to a place where there is no scratch or foreign matter is present. The amount of light only changes slightly from the amount of incident light.
[0103]
On the other hand, when light is irradiated on a spot on a photographic film, a part of the irradiated light is refracted by the scratch, so that the amount of transmitted light when the light is irradiated on the spot with the scratch ( Attenuation of light incident on the photographic film by adding attenuation due to light refraction due to scratches to the attenuation due to light absorption by the photographic film. Attenuate by the amount. Although FIG. 8A shows a case where the light incident side is scratched, the same applies to the case where the light emission side is scratched.
[0104]
Since refraction of light due to scratches occurs even with IR light, the amount of IR light transmitted when IR light is irradiated to the scratched part is attenuated by an attenuation amount corresponding to attenuation due to light refraction due to scratches. To do. As shown in FIG. 8B as an example, the refraction of light due to scratches becomes more noticeable as the scratch scale (depth, etc.) increases (same for visible light and IR light). The amount of transmitted light when IR light is applied to a scratched portion decreases as the size of the scratch increases. Accordingly, it is possible to detect the scale of the scratch on the photographic film based on the attenuation amount of the transmitted light amount of the IR light.
[0105]
In addition, when light is irradiated on a part of a photographic film where foreign matter such as dust is attached, the irradiated light is reflected by the foreign matter, and the foreign matter depends on the size and type (light transmittance) of the foreign matter. The amount of transmitted light when light is irradiated on the part marked with is greatly attenuated by the foreign matter. The attenuation of the amount of transmitted light when light is applied to a part with a foreign object is the same when IR light is applied to the part.
[0106]
As described above, when the IR light is transmitted through the photographic film, the amount of transmitted light changes only at a spot on the photographic film where there are scratches or foreign matter, and even if an image is recorded on the photographic film, the image Therefore, it is possible to detect scratches and foreign matter attached to the photographic film by irradiating the photographic film with IR light and detecting the amount of transmitted light.
[0107]
Based on the above, in step 120, the defect detection process is performed as follows. As described above, the amount of transmitted light when irradiating IR light on a photographic film is generally constant regardless of the position on the image as described above, and decreases only at a point where the photographic film is scratched or foreign (see FIG. 9). ). Since the IR data represents the amount of IR light transmitted at each position on the image to be processed, the amount of IR light transmitted (for example, transmitted light) represented by the IR data at a location on the image to be processed without scratches or foreign matter. The maximum value of the light amount is used as a reference value. Then, the amount of transmitted IR light is compared with a reference value for each pixel, and the amount of change (decrease amount) in the amount of transmitted light with respect to the reference value is a predetermined value (a little of the amount of transmitted IR light at a place where there is no scratch or foreign object) All the pixels having a value equal to or greater than the value determined in consideration of the fluctuations in (1) are detected as defective pixels belonging to the defect portion to be corrected.
[0108]
In step 120, the detected defective pixels are classified into defective pixels belonging to the same defective portion based on the positional relationship between the defective pixels (for example, whether they are adjacent to each other), and information about each defective portion ( For example, information indicating defective pixels belonging to each defective portion and information such as a decrease in the amount of IR light transmitted through each defective pixel are stored in the RAM 48 or the like.
[0109]
  The defect detection process described above is performed using R, G, B, and IR data that have undergone various corrections, so that only defective parts corresponding to scratches and foreign matters on photographic film can be detected reliably and accurately. In addition, it is possible to prevent erroneous detection of the range of the defective portion. Step 120 claims7This corresponds to the defect detection means described in (1).
[0110]
In the next step 122, a correction value for correcting each defect portion is calculated for each defect portion to be corrected detected by the defect portion detection process. First, the principle of defect repair will be described.
[0111]
As shown in FIG. 8B, the emulsion layer of the photographic film includes R, G, and B photosensitive layers, and the photographic film (negative film) on which the image has been recorded by exposure and developed. Film) has a C negative image formed on the R photosensitive layer, an M negative image formed on the G photosensitive layer, and a Y negative image formed on the B photosensitive layer. Of the visible light transmitted through the photographic film, the R light is attenuated (absorbed) in the R photosensitive layer by an attenuation amount corresponding to the transmission density of the C negative image, and the G negative image is absorbed in the G photosensitive layer. The B light is attenuated (absorbed) by an attenuation amount corresponding to the transmission density of the negative image of Y in the B photosensitive layer.
[0112]
Here, as shown in FIG. 8B as an example, when the back surface opposite to the emulsion surface is scratched, the ratio of light absorption in the R, G, B photosensitive layers to the transmitted light is as follows. The same as when there is no scratch. That is, in FIG. 8B, the amount of light incident on the photographic film is expressed as I.₀, The amount of transmitted R light, G light, and B light when there is no scratch,_0R, I_0G, I_0BThe amount of light incident on the emulsion layer through a straight line through the scratched part when the scratch is made is I₁(I₁<I₀: I₀-I₁Is the amount of transmitted light of R light, G light, and B light when there is a flaw._1R, I_1G, I_1BThen, the relationship of the following formula (2) is established.
I_0R/ I₀≒ I_1R/ I₁  I_0G/ I₀≒ I_1G/ I₁  I_0B/ I₀≒ I_1B/ I₁      ... (2)
[0113]
Therefore, the defect portion corresponding to the scratched portion on the back surface changes only in luminance as compared with the case where there is no scratch, and the color information of the image recorded on the photographic film is stored. Therefore, the defect portion of the image represented by the image data can be corrected by applying the luminance adjustment method and adjusting the luminance of the defect portion area.
[0114]
On the other hand, when the emulsion surface is scratched as shown in FIG. 8C as an example, if the scratch is shallow, a part of each photosensitive layer is scraped, so that R, The ratio of light absorption in each of the photosensitive layers G and B varies from that in the case where there is no scratch. Further, if the scratches are very deep such that all the photosensitive layers are peeled off, light absorption in each photosensitive layer with respect to transmitted light does not occur. Therefore, in any case, the relationship of the expression (2) is not established.
[0115]
In this way, the defect portion corresponding to the scratched portion of the emulsion surface changes in brightness and color as compared with the case where there is no scratch regardless of the depth of the scratch. Since the color information of the image recorded in the image is also lost, it is difficult to correct the defective portion with high accuracy even if the luminance is adjusted. For this reason, a correction method (interpolation method) that determines the brightness and density of the defective portion by interpolation from information on the area around the defective portion is suitable for correcting the defective portion corresponding to the scratched portion of the emulsion surface. ing. In addition, since the brightness and the color of the defective portion caused by the foreign matter attached to the photographic film change as compared with the case where the foreign matter is not attached, when correcting the defective portion described above, The interpolation method is also suitable.
[0116]
In step 122, first, predetermined feature amounts for determining whether to correct each defect portion to be corrected by applying an interpolation method or applying a luminance adjustment method are calculated. In the present embodiment, as an example of the predetermined feature amount, a feature amount representing a correlation between changes in the amount of transmitted light of R light, G light, and B light in the defective portion is used.
[0117]
For example, when the back surface of a photographic film is scratched, as shown in FIG. 9 (A) as an example, the amount of transmitted light of R light, G light, and B light changes in substantially the same manner at the scratched portion. As shown, the correlation between the changes in the amount of transmitted light of R light, G light, and B light is high. On the other hand, when the emulsion surface of the photographic film is scratched, as shown in FIG. 9B as an example, the change in the amount of transmitted R light, G light, and B light at the scratched portion is not constant. , R light, G light, and B light have a low correlation of changes in the amount of transmitted light (the same applies when foreign matter is attached to the photographic film).
[0118]
FIG. 9 shows a typical case, and there are many cases where it is not clear which correction method should be applied, such as scratches on both sides of the photographic film. However, it is preferable to apply the luminance adjustment method if information about the color of the subject remains in the R, G, B image data, and it is preferable to apply the interpolation method when the information does not remain. Predetermined feature quantity (for example, a value obtained by integrating the difference between the differential values of the changes in the transmitted light amount of R light, G light, and B light) representing the correlation of the change in the transmitted light amount of R light, G light, and B light in the defective portion. Based on this, it is possible to appropriately determine the correction method to be applied for each defective portion.
[0119]
When each of the predetermined feature amounts described above is calculated for each defect part to be corrected, the setting value of the application range of both correction methods (a threshold value indicating the boundary of the application range of both correction methods) is fetched, and each defect part Are compared with the set values to determine whether to perform correction by applying an interpolation method or by applying a luminance adjustment method for each defective portion as a unit.
[0120]
Then, for the defective portion determined to be corrected by applying the interpolation method, the correction value is calculated by applying the interpolation method. That is, the brightness and color of the defect part to be corrected are newly calculated by interpolation from the brightness and color of the area around the defect part, and the values (R, G, and B) of the pixels in the defect part obtained by the interpolation calculation are calculated. The density value may be a value representing hue, lightness, or saturation), and D1 is the original value of each pixel, D2 is the correction level, and α is the correction value D3 of each pixel in the defective portion. Ask according to.
D3 = α · D1 + (1−α) D2 (3)
The above processing is performed for each defective portion determined to be corrected by applying the interpolation method, and a correction value for each defective portion is obtained.
[0121]
  In addition, for the defective portion determined to be corrected by applying the luminance adjustment method, the correction value is calculated by applying the luminance adjustment method. That is, the luminance correction amount of the defective portion is calculated based on the amount of change in the amount of transmitted IR light in the defect portion to be corrected, and the luminance value of each pixel in the defective portion corrected according to the luminance correction amount is L1, each pixel Let L2 be the original luminance value, and α be the current set value of the “defect portion correction degree”, and obtain the correction value (luminance value) L3 of each pixel in the defect portion according to equation (4).
L3 = α · L1 + (1−α) L2 (4)
The above processing is performed for each defective portion determined to be corrected by applying the brightness adjustment method, and a correction value for each defective portion is obtained. The above-mentioned step 122 is claimed.7This corresponds to the correction means described in (1).
[0122]
When the correction values are calculated for all the defect portions to be corrected, the process proceeds to step 124, and the correction value for each defect portion is changed to information indicating the position of the defect portion (for example, the defect pixel constituting each defect portion). Address) and the image processor 40, and the defect portion correction value determination process is terminated.
[0123]
  The image processor 40 receives the image data that has undergone the corrections in steps 102 to 118 in the defect portion correction value determination process (the correction may be performed by the image processor 40). On the other hand, the defect portion correction value determination process is performed by the control unit 42 to correct the defect portion in accordance with the correction value notified from the control unit 42 (specifically, the notified correction value of each defective pixel belonging to the defect portion). A defective portion correction process (replaced with a pixel value) is performed. Thereby, all the defective parts to be corrected are automatically corrected. Thus, the image processor 40 is also claimed.7This corresponds to the correction means described in (1).
[0124]
Further, the image processor 40 performs various image processing on the image data that has been subjected to the defect portion correction processing under the processing conditions determined by the setup calculation in the control unit 42, and causes the I / O controller 38 and the I / F circuit 54 to operate. Output to the printer 16. As a result, the defective portion selected as the correction target is erased from the image recorded by exposure on the photographic paper 68.
[0125]
In the above description, the example using the film scanner 12 configured to perform reading (detection of transmitted light amount) for each of the R, G, B, and IR wavelength ranges by the single area CCD 30 has been described. However, a film scanner having a configuration in which reading for each wavelength region is performed by two or more photoelectric conversion elements may be used. As an example, FIG. 10 shows a film scanner 72 having a configuration in which reading for each wavelength band is performed by different line sensors 70B, 70G, 70R, and 70IR.
[0126]
In this film scanner 72, the light emitted from the light source 20 is shaped into a long slit-shaped light beam along the width direction of the photographic film 26 by the light diffusion box 74, and is irradiated onto the photographic film 26 and transmitted through the photographic film 26. The light is incident on each of the line sensors 70B, 70G, 70R, and 70IR via the imaging lens 28. The light receiving surface of each line sensor is provided with a filter that transmits only the light in the wavelength range to be detected among the incident light, and the photographic film 26 is conveyed at a constant speed. Reading about the wavelength range (detection of the amount of transmitted light of each wavelength range) is performed.
[0127]
As in the case of the film scanner 72 described above, when reading with respect to each wavelength region is performed with a plurality of photoelectric conversion elements, each of the respective wavelength regions is obtained by reading due to the difference in the arrangement position of each photoelectric conversion element. Pixel shift occurs in image information. For example, in the film scanner 72, each reading position of the adjacent line sensors on the photographic film 26 is n times (n is an integer) the distance that the photographic film 26 is transported per reading cycle. If the line sensors are respectively arranged, pixel deviation does not occur if the reading timing of each line sensor is shifted by n cycles, but the interval between the reading positions is set to the one cycle due to, for example, an error in the mounting position of the line sensor. If it is a non-integer multiple of the hit film transport distance, the aforementioned pixel shift occurs.
[0128]
In such a case, the amount of misalignment with respect to the intended position of the reading position of each line sensor may be measured and stored in advance, and misalignment correction may be performed. This misalignment correction is performed by, for example, correcting the misalignment by the measured amount of misalignment in distortion aberration correction and lateral chromatic aberration correction in a direction parallel to the conveyance direction of the photographic film (for example, the x direction in FIG. 5A). This can be realized by correcting so that the positions of all the pixels move in the direction opposite to the direction. Thereby, the shift of the pixel position due to the difference in the arrangement position of the photoelectric conversion elements can be corrected.
[0129]
  The above-mentioned correction is claimed5The line sensors 70B, 70G, and 70R of the film scanner 72 are claimed in the claims.5The line sensor 70IR is included in the first photoelectric conversion element according to claim 1.5This corresponds to the second photoelectric conversion element described in 1).
[0130]
In addition, the interpolation method and the luminance adjustment method have been described above as an example of the defect portion correction method, but the present invention is not limited to this, and a so-called blurring method that blurs the defect portion by applying a low-pass filter or the like is applied. Also good.
[0131]
Moreover, although the structure which reads an image by photoelectrically converting the light which permeate | transmitted the photographic film was demonstrated above, it is not limited to this, The structure which reads an image by photoelectrically converting the light which reflected the photographic film May be adopted. Needless to say, the image recording material is not limited to a photographic film, and a photographic material other than the photographic film, plain paper, an OHP sheet, or the like may be used as the image recording material.
[0132]
Further, in the above description, an example in which R, G, B reading is performed during pre-scanning and R, G, B, IR reading is performed during fine scanning has been described. However, the present invention is not limited to this. It may be performed only at the time of canning, or may be performed at the time of pre-scanning and fine scanning.
[0133]
【The invention's effect】
  As explained aboveBookThe inventionVisible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting visible light transmitted or reflected through the image recording area, and light in a wavelength region including a part of the visible region is invisible. The difference between each image information attributed to the optical characteristics is visible to at least one of the invisible image information obtained by irradiating the image recording area as light and detecting the invisible light transmitted or reflected through the image recording area. Corrects the change in the amount of detected invisible light caused by light in the wavelength region including part of the wavelength being irradiated as invisible lightBecause I tried to,Lack ofIt has an excellent effect that it is possible to improve the correction accuracy of the recessed portion.
[0134]
  Also bookThe inventionVisible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected through the image recording area by the first detection means, and invisible light on the image recording area The image recording is performed as a difference between the image information caused by optical characteristics with respect to at least one of the invisible image information obtained by detecting the invisible light that has been irradiated and transmitted or reflected through the image recording area by the second detection unit. Corrects the difference in the amount of light detected by each detection means due to the wavelength dependence of the attenuation of the light transmitted through or reflected by the image recording material.Because I tried to,Lack ofIt has an excellent effect that it is possible to improve the correction accuracy of the recessed portion.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of an image processing system according to an embodiment.
FIG. 2 is a perspective view showing a schematic configuration of a film scanner.
FIG. 3 is a diagram showing an example of spectral characteristics of light irradiated on a photographic film.
FIG. 4 is a flowchart showing the content of a defect portion correction value determination process.
5A is an xy coordinate system set for an image, and FIG. 5B is x for explaining distortion aberration correction and lateral chromatic aberration correction._Py_PIt is a conceptual diagram which shows a coordinate system.
FIG. 6 is a diagram showing an example of the relationship between the wavelength of light transmitted through a photographic film and the degree of attenuation with respect to transmitted light.
7A shows an example of response characteristics of image data Q and unsharp mask image data QUS, FIG. 7B shows an example of response characteristics of image data (Q-QUS), and FIG. 7C shows image data Q and sharpness. It is a diagram which shows an example of the response characteristic of the image data QL after degree correction, respectively.
FIG. 8A is a conceptual diagram showing light transmission in a photographic film where there are no scratches and foreign matter, a scratched location, and a foreign matter location, and FIG. When the back surface is scratched, (C) is a conceptual diagram showing light transmission when the emulsion surface of the photographic film is scratched.
FIGS. 9A and 9B show examples of changes in the amount of transmitted R light, G light, B light, and IR light when the back surface is scratched, and FIG. 9B is when the emulsion surface is scratched. FIG.
FIG. 10 is a schematic configuration diagram of a film scanner according to another embodiment of the present invention.
[Explanation of symbols]
12 Film scanner
14 Image processing device
20 Light source
23 Filter unit
26 photographic film
28 Imaging lens
30 area CCD
40 Image processor
42 Control unit
72 Information storage media

Claims (11)

Visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and a wavelength partially including the visible area acquiring for acquiring invisible image information obtained by detecting the non-visible light transmitted through or reflected by the image recording area shines light of the said image recording area light as invisible light range by the second detecting means Means,
As at least one of the visible image information and the invisible image information , light in a wavelength region including a part of the visible region is irradiated as invisible light as a difference between the image information due to optical characteristics. Correction means for correcting the change in the detected light amount of the invisible light by the second detection means ,
An image processing apparatus. Visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and invisible to the image recording area An acquisition means for acquiring invisible image information obtained by irradiating light and detecting the invisible light transmitted or reflected by the image recording area by the second detection means;
As a difference between the image information due to optical characteristics with respect to at least one of the visible image information and the non-visible image information, the wavelength dependence of the attenuation by the image recording material of the light transmitted or reflected by the image recording material Correction means for correcting the difference in the amount of light detected by each of the detection means due to
An image processing apparatus. The acquisition means is the visible image for each of the plurality of wavelength ranges obtained by the first detection means detecting the visible light transmitted or reflected through the image recording area for each of the plurality of wavelength ranges as the visible image information. Acquire each image information,
The said correction | amendment means correct | amends the difference between each said image information resulting from an optical characteristic with respect to the visible image information and the said invisible image information for every said several wavelength range. 2. The image processing apparatus according to 2 . The first detection means and the second detection means detect visible light or invisible light that is transmitted or reflected through an image recording region of the image recording material and imaged by an imaging lens,
Said correction means, wherein a difference in the image information, the image processing apparatus according to claim 1 or claim 2, wherein also the deviation of the pixel position due to magnification chromatic aberration or distortion aberration and corrects the imaging lens . The first detection means detects visible light with a first photoelectric conversion element, and the second detection means detects invisible light with a second photoelectric conversion element different from the first photoelectric conversion element,
The correction means also corrects a shift in pixel position due to a difference in arrangement position between the first photoelectric conversion element and the second photoelectric conversion element as a difference between the image information. The image processing apparatus according to claim 1 or 2 . The first detection means and the second detection means detect visible light or invisible light that is transmitted or reflected through an image recording region of the image recording material and imaged by an imaging lens,
The correction means also corrects a difference in sharpness of an image represented by each image information due to a wavelength dependency of a focal length of the imaging lens as a difference between the image information. Alternatively, the image processing apparatus according to claim 2 . Based on the visible image information and non-visible image information that have undergone correction by the correction means, a defect portion detection means that detects a defect portion of the image represented by the visible image information;
Correction means for correcting the defective portion detected by the detection means for the visible image information;
The image processing apparatus according to claim 1 or claim 2, wherein further comprising a. The image recording area of the image recording material is irradiated with visible light to transmit the image recording area. Visible image information obtained by detecting excessive or reflected visible light by the first detection means, and light in a wavelength region partially including the visible region is irradiated to the image recording region as the invisible light, and the image recording is performed. Obtaining invisible image information obtained by detecting the invisible light transmitted or reflected by the region by the second detection means;
As at least one of the visible image information and the invisible image information, light in a wavelength region including a part of the visible region is irradiated as invisible light as a difference between the image information due to optical characteristics. Correcting the change in the amount of invisible light detected by the second detection means
Image processing method. Visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and invisible light on the image recording area To obtain invisible image information obtained by detecting the invisible light transmitted or reflected by the image recording area by the second detection means,
As a difference between the image information due to optical characteristics with respect to at least one of the visible image information and the non-visible image information, the wavelength dependence of the attenuation by the image recording material of the light transmitted or reflected by the image recording material An image processing method for correcting a difference in the amount of light detected by each of the detection means due to the above . Computer
Visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and a wavelength partially including the visible area acquiring for acquiring invisible image information obtained by detecting the non-visible light transmitted through or reflected by the image recording area shines light of the said image recording area light as invisible light range by the second detecting means means,
In addition, at least one of the visible image information and the invisible image information is irradiated with light in a wavelength region including part of the visible region as invisible light as a difference between the image information due to optical characteristics. Correction means for correcting a change in the amount of invisible light detected by the second detection means
A recording medium on which a program for functioning as a recording medium is recorded. Computer
Visible image information obtained by irradiating the image recording area of the image recording material with visible light and detecting the visible light transmitted or reflected by the image recording area by the first detection means, and invisible light on the image recording area Acquisition means for acquiring non-visible image information obtained by detecting by the second detection means non-visible light that is transmitted through or reflected by the image recording area,
The wavelength of the attenuation by the image recording material of light transmitted or reflected by the image recording material as a difference between the image information due to optical characteristics with respect to at least one of the visible image information and the invisible image information Correction means for correcting the difference in the amount of light detected by each of the detection means due to dependency
A recording medium on which a program for functioning as a recording medium is recorded.

Images