JP4013456B2 - Imaging device - Google Patents

Imaging device Download PDF

Info

Publication number
JP4013456B2
JP4013456B2 JP2000206465A JP2000206465A JP4013456B2 JP 4013456 B2 JP4013456 B2 JP 4013456B2 JP 2000206465 A JP2000206465 A JP 2000206465A JP 2000206465 A JP2000206465 A JP 2000206465A JP 4013456 B2 JP4013456 B2 JP 4013456B2
Authority
JP
Japan
Prior art keywords
value
average
weighted
area
error
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
JP2000206465A
Other languages
Japanese (ja)
Other versions
JP2002023216A (en
Inventor
彰浩 田村
靖利 山本
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Corp
Panasonic Holdings Corp
Original Assignee
Panasonic Corp
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Panasonic Corp, Matsushita Electric Industrial Co Ltd filed Critical Panasonic Corp
Priority to JP2000206465A priority Critical patent/JP4013456B2/en
Publication of JP2002023216A publication Critical patent/JP2002023216A/en
Application granted granted Critical
Publication of JP4013456B2 publication Critical patent/JP4013456B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Landscapes

  • Exposure Control For Cameras (AREA)
  • Studio Devices (AREA)

Description

【０００１】
【発明の属する技術分野】
本発明は、逆光判定を行い判定結果により重み付け平均値による制御を連続的に切り換える事により適切な露光制御を行うことができる撮像装置に関する。
【０００２】
【従来の技術】
従来の撮像装置は、特開平６−３１１４２３号公報に記載されたものが知られている。
【０００３】
図１０は同公報に示されている従来の撮像装置のブロック図である。図１０において、１００１が光学系、１００２が絞り、１００３が撮像素子、１００４がアンプ、１００５が信号処理回路、１００６が出力端子、１００７が全画面平均検出部、１００８が絞り制御部、１００９が第1・第２領域輝度比較部、１０１０がピーク値検出部、１０１１がピーク値誤差算出部、１０１２が平均値誤差算出部、１０１３が加重平均誤差算出部、１０１４が重み算出部である。
【０００４】
以上のように構成された従来の撮像装置の動作について説明する。撮像素子１００３の映像信号から、第１・第２領域輝度比較部１００９で上下輝度差ｐを求める。ピーク値誤差算出部１０１１で、ピーク値検出部１０１０からのピーク値と目標値とのピーク値誤差を算出する。平均値誤差算出部１０１２で、全画面平均検出部１００７からの全画面平均と目標値との平均値誤差を算出する。重み算出部１０１４でｐよりピ−ク値誤差と平均値誤差に対する重み係数を算出する。加重平均誤差算出部１０１３で、ピーク値誤差算出部１０１１からのピ−ク値誤差と平均値誤差算出部１０１２からの平均値誤差と、重み算出部１０１４からの重み係数を用いて加重平均誤差を計算する。絞り制御部１００８で前記加重平均誤差を０に保つ様に絞り開度を制御する。
【０００５】
第１・第２領域輝度比較部１００９で上下輝度差ｐを求めることで、逆光、過順光の判定を行う。画面上部を含む第１の領域の平均輝度が画面中央を含む第２の領域の平均輝度より大きい時に逆光と判定する。この時、重み算出部１０１４でｐより平均値誤差に対するピ−ク値誤差の重み係数を更に小さくする。平均値誤差は絞り１００２が開く方向に、ピーク値誤差は絞り１００２が閉じる方向に働くので、ピーク値誤差の重み係数が小さくなると絞り１００２が開くようになる。従って、逆光被写体を撮影する時は逆光被写体が暗くならないように絞りを開き、適切な露光制御を行うことができる。
【０００６】
【発明が解決しようとする課題】
このような従来の撮像装置においては、逆光、過順光の判定を行い、適切な露光制御を行うことが要求されている。
【０００７】
しかしながら、上記の従来例においては、強逆光時、逆光の判定することはできるが、平均値誤差に対するピーク値誤差の重み係数を小さくするだけでは、逆光被写体が明るくなるような十分な逆光補正を行うことができないという課題がある。
【０００８】
また、上記の従来例においては、目標値を上げることによって強逆光時も十分な逆光補正を行うことができるが、通常の順光被写体で絞りが開きすぎてしまうという課題もある。
【０００９】
また、上記の従来例においては、過順光の時、全画面平均が小さくなり、過順光判定されると、平均値誤差の重みが大きくなるので、絞りが開きすぎるという課題もある。
【００１０】
本発明は、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することを目的とする。
【００１１】
【課題を解決するための手段】
本発明の第１の観点の撮像装置は、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、
全画面領域内のピーク輝度値を検出するピーク値１検出部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値１検出部で得られたピーク輝度値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えたものである。
【００１２】
第１の観点の撮像装置によれば、第１・第２領域輝度比較部で画面上部を含む第１の領域の平均輝度と、画面中央を含む第２の領域の平均輝度を比較することで逆光判定を行う。この比較結果によって、平均値誤差とピーク値誤差の適当な加重平均の重みを算出し、更に重み付け１全画面平均値と重み付け２全画面平均値の適当な加重平均の重みを算出することで、絞りを更に開くことができるので、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【００１３】
第２の観点の撮像装置は、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の各重みが略均等の重み付けに設定された代表輝度値と第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、全画面領域内のピーク１輝度値を検出するピーク値１検出部と、画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えたものである。
【００１４】
第２の観点の撮像装置によれば、第１・第２領域輝度比較部で画面上部を含む第１の領域の平均輝度と、画面中央を含む第２の領域の平均輝度を比較することで逆光判定を行う。この比較結果によって、平均値誤差とピーク値誤差の適当な加重平均の重みを算出し、更に重み付け１全画面平均値と重み付け２全画面平均値の適当な加重平均の重みを算出することで、絞りを更に開くことができる。更にピーク１輝度値とピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができるので、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【００１５】
第３の観点の撮像装置は、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、全画面領域内のピーク１輝度値を検出するピーク値１検出部と、画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、前記重み付け２全画面平均算出部で得られた重み付け２全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えたものである。
【００１６】
第３の観点の撮像装置によれば、第１・第２領域輝度比較部で画面上部を含む第１の領域の平均輝度と、画面中央を含む第２の領域の平均輝度を比較することで逆光判定を行う。この比較結果によって、平均値誤差とピーク値誤差の適当な加重平均の重みを算出することで、絞りを開くことができ、更にピーク１輝度値とピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができるので、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【００１７】
第４の観点の撮像装置は、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記平均値誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部とを備えたものである。
【００１８】
第４の観点の撮像装置によれば、第１・第２領域輝度比較部で画面上部を含む第１の領域の平均輝度と、画面中央を含む第２の領域の平均輝度を比較することで逆光判定を行う。この比較結果によって、重み付け１全画面平均値と重み付け２全画面平均値の適当な加重平均の重みを算出することで、絞りを更に開くことができるので、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【００１９】
【発明の実施の形態】
以下、本発明の実施の形態について、図１ないし図９を用いて説明する。
【００２０】
（実施の形態１）
本発明の第１の実施例について、図１ないし図６を参照しながら説明する。
【００２１】
図１は、本発明の実施の形態１による撮像装置のブロック図を示す。図１において、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０２の出力は、ダブルサンプリング、利得調整された後、アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調等の信号処理を行い出力端子１０６から映像信号を出力する。
【００２２】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０の入力端に接続される。領域輝度検出部１０７では、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する。ピーク値１検出部１１０では、全画面領域内の最大輝度信号の値をピーク値として検出する。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較する。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２とピーク重み算出部１１６の入力端に接続される。平均重み算出部１１２では、平均重みを算出する。ピーク重み算出部１１６では、ピーク重みを算出する。また、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９では、それぞれ重み付け１全画面平均値と重み付け２全画面平均値を算出する。重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。全画面平均値演算部１１３では、全画面平均値を演算する。全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、平均値誤差を算出する。また、ピーク値１検出部１１０の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、ピーク値誤差を算出する。ピーク値誤差算出部１１４と平均値誤差算出部１１５とピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。加重平均誤差算出部１１７では加重平均誤差を算出する。加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【００２３】
図２は本実施の形態の撮像装置における領域輝度検出部の全画面領域の領域分割を示す模式図である。図２において、２０１は第１領域、２０２は第２領域を示す。
【００２４】
図３は本実施の形態の撮像装置における重み付け１全画面平均算出部と重み付け２全画面平均算出部の重み係数を示す構成図である。図３において、（ａ）は重み付け１全画面平均値の重み係数、（ｂ）は重み付け２全画面平均値の重み係数、（ｃ）は重み付け全画面平均値の重み係数の例１、（ｄ）は重み付け全画面平均値の重み係数の例２を示す。
【００２５】
図４は本実施の形態の撮像装置におけるピーク値１検出部のピーク検出領域を示す検出領域図である。図４において、２０１は第１領域、２０２は第２領域、４０１は第１領域と第２領域をあわせた全画面領域を示す。
【００２６】
図５は本実施の形態の撮像装置における全画面平均値演算部で全画面平均値を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【００２７】
図６は本実施の形態の撮像装置における加重平均誤差算出部で加重平均誤差を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【００２８】
以上のように構成された本実施の形態の撮像装置について、以下その動作を説明する。図１において、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０３の出力は、ダブルサンプリング、利得調整された後、アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調などの信号処理をして出力端子１０６から映像信号を出力する。
【００２９】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０の入力端にも接続される。
【００３０】
図２を用いて、領域輝度検出部１０７の動作説明を行う。領域輝度検出部１０７では、図２に示したように全画面領域を水平方向に８領域に分割し、垂直方向に６領域に分割する。各領域内において画素毎に輝度信号の単純加算を行い、任意の数値（例えば、２５６）で割ることによって各領域の代表輝度値を検出する。任意の数値を各領域内の画素数にすれば、代表輝度値は各領域の平均輝度値になる。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。
【００３１】
図２を用いて、第１・第２領域輝度比較部１１１の動作説明を行う。第１・第２領域輝度比較部１１１では、図２に示した画面上部を含む第１領域内の各領域の代表輝度値を平均した平均輝度から、画面中央を含む第２領域内の各領域の代表輝度値を平均した平均輝度を減算し、第1領域と第２領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を算出する。輝度差分ＣＮＴの値が負になった場合は、負クリップを行う。また、輝度差分ＣＮＴの値が最大値を越えた場合は最大値クリップを行う。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２とピーク重み算出部１１６の入力端に接続される。
【００３２】
図３を用いて、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９の動作説明を行う。重み付け１全画面平均算出部１０８では、図３（ａ）に示した重み付け１全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け１全画面平均値を算出する。
【００３３】
同様に、図３（ｂ）に示した重み付け２全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け２全画面平均値を算出する。
【００３４】
図３（ｂ）に示した重み付け２全画面平均値の重み係数は画面上部になるに従い、重みが小さくなる設定になっている。画面上部に明るく輝度信号が大きい被写体がある場合は、重み付け２全画面平均値を上記の方式で算出すると、画面上部の重みが小さくなっているため、全画面の重みが均一になっている重み付け１全画面平均値より小さくなる。また逆に、画面上部に暗く輝度信号が小さい被写体がある場合は、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。
【００３５】
重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。
【００３６】
図４を用いて、ピーク値１検出部１１０の動作説明を行う。ピーク値１検出部１１０では、図４に示した全画面領域内の最大輝度信号の値をピーク値として検出する。
【００３７】
図５を用いて、平均重み算出部１１２の動作説明を行う。平均重み算出部１１２では、図５に示した第１領域と第２領域の輝度差分と平均重みの関係になるように平均重みαを算出する。平均重みαは、重み付け２全画面平均の重み係数を指し、重み付け１全画面平均の重み係数は、１００％−αで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、重み付け１全画面平均が１００％になるように平均重みαを０％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、重み付け２全画面平均が１００％になるように平均重みαを１００％にする。
【００３８】
図５を用いて、全画面平均値演算部１１３の動作説明を行う。全画面平均値演算部１１３では、重み付け１全画面平均値と重み付け２全画面平均値を平均重みαを用いて加重平均を行い、全画面平均値を算出する。全画面平均値をＡＶＥＹ、重み付け１全画面平均値をＡＶＥＹ１、重み付け２全画面平均値をＡＶＥＹ２とすると、ＡＶＥＹ＝（ＡＶＥＹ１×（１００−α）＋ＡＶＥＹ２×α）／１００、で示す式によって、全画面平均値を算出することができる。
【００３９】
全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、所定の目標値から全画面平均値を減算して平均値誤差を算出する。
【００４０】
また、ピーク値１検出部１１０の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、所定の目標値からピーク値を減算してピーク値誤差を算出する。
【００４１】
ここで、絞り１０２の開閉動作が安定した状態では通常、ピーク値誤差は負の値になり、平均値誤差は正の値になる。ピーク値誤差と平均値誤差を加重平均し打ち消しあって加重平均誤差が０になる絞り位置で絞り１０２が釣り合うようになっている。ピーク値誤差算出部１１４と平均値誤差算出部１１５の出力端は加重平均誤差算出部１１７の入力端に接続される。
【００４２】
図６を用いて、ピーク重み算出部１１６の動作説明を行う。ピーク重み算出部１１６では、図６に示した第１領域と第２領域の輝度差分とピーク重みの関係になるようにピーク重みβを算出する。ピーク重みβは、ピーク値誤差の重み係数を指し、平均値誤差の重み係数は、１００％−βで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、ピーク値誤差が２５％になるようにピーク重みβを２５％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、平均値誤差が１００％になるようにピーク重みβを０％にする。ピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。
【００４３】
図６を用いて、加重平均誤差算出部１１７の動作説明を行う。加重平均誤差算出部１１７では、ピーク値誤差算出部１１４で得られたピーク値誤差と平均値誤差算出部１１５で得られた平均値誤差をピーク重み算出部１１６で得られたピーク重みβを用いて加重平均を行い、加重平均誤差を算出する。加重平均誤差をＧＯＳＡ、ピーク値誤差をＰＧＯＳＡ、平均値誤差をＡＧＯＳＡとすると、ＧＯＳＡ＝（ＡＧＯＳＡ×（１００−β）＋ＰＧＯＳＡ×β）／１００、で示す式で加重平均誤差を算出することができる。
【００４４】
加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【００４５】
逆光判定の動作について、図１ないし図６を用いて動作説明を行う。逆光判定は、第１・第２領域輝度比較部１１１で行う。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度から、画面中央を含む第２領域の平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを演算する。逆光時に人などの主被写体を画面中央にして撮影すると、主被写体は暗くなり、画面上部には空や雲など明るい被写体になることが多いので、第1領域と第２領域の輝度差分ＣＮＴは正の値になる。強逆光になるほど、第1領域と第２領域の輝度差分ＣＮＴの値は大きくなる。すなわち、第1領域と第２領域の輝度差分ＣＮＴが正になるか負になるかで逆光判定を行い、輝度差分ＣＮＴの値の大きさで逆光の強度を判断する。また、過順光時に主被写体を画面中央にして撮影すると、主被写体が明るくなっているので、第1領域と第２領域の輝度差分ＣＮＴは負の値になる。
【００４６】
次に、逆光判定された時の逆光補正の動作について、図１ないし図６を用いて動作説明を行う。逆光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが正の値になるので、第２領域に比べ第１領域に明るい被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より小さくなる。
【００４７】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが大きな値になっていれば、平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け２全画面平均に近い全画面平均値を算出する。よって、全画面平均値は従来の撮像装置の全画面平均値より小さくなる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の全画面平均値より小さくなっているので、平均値誤差は大きくなる。
【００４８】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。従って、逆光時は、ピーク重みβは小さくなる。よって、加重平均誤差算出部１１７で算出される加重平均誤差は平均値誤差に近い値になる。平均値誤差が大きくなっているので加重平均誤差も大きくなる。加重平均誤差が大きいと、加重平均誤差を小さくするように、絞り１０２を開く方向に絞り制御部１１８が働くので、逆光時は従来の撮像装置より絞り１０２を開き逆光補正を行うことができる。
【００４９】
次に、順光・過順光判定された時の絞り制御の動作について、図１ないし図６を用いて動作説明を行う。順光・過順光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になるので、画面中央を含む第２領域に比べ画面上部を含む第１領域に暗い被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。また、重み付け１全画面平均値の重み係数は均等なっているので、重み付け１全画面平均値は従来の撮像装置の全画面平均値と等しくなる。
【００５０】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になっていれば、負クリップされる。平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが０の時は、重み付け１全画面平均の重みが１００％になるように平均重みを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け１全画面平均を全画面平均値として算出する。よって、全画面平均値は従来の撮像装置の全画面平均値と同じになる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の撮像装置の全画面平均値と同じなので、平均値誤差も等しくなる。
【００５１】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが負クリップされた０になるので、ピークの重み係数が最大値になるようにピーク重みを算出する。従って、順光・過順光時は、ピーク重みは２５％になる。よって、加重平均誤差算出部１１７では平均値誤差とピーク値誤差を３：１で加重平均を行い加重平均誤差を算出する。
【００５２】
ピーク重みの最大値を従来撮像装置の設定と等しく設定すれば、ピーク値誤差と平均値誤差が従来の撮像装置と等しいので、加重平均誤差も従来の撮像装置と等しくなる。よって、順光・過順光時は従来の撮像装置の絞り制御と同等の制御を行うことができる。
【００５３】
従って、逆光時は従来の撮像装置より絞りを開くことができ、順光・過順光時は従来の撮像装置と同等の絞り制御になるので、適切な露光制御を行うことができる撮像装置を得ることができる。
【００５４】
このように構成することによって、第１・第２領域輝度比較部１１１で画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較することで逆光判定を行う。逆光判定時は、平均値誤差の重み係数がピーク値誤差の重み係数より大きくなるように加重平均のピーク重みを算出し、更に重み付け１全画面平均値の重み係数が重み付け２全画面平均値の重み係数より小さくなるように加重平均の平均重みを算出することで、絞りを更に開くことができる。更に、順光判定時は、従来の撮像装置と同等の絞り制御ができるので、順光被写体や強逆光被写体に対しても、逆光、順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果を得られる。
【００５５】
なお、上記動作説明において全画面の領域分割を水平８領域分割、垂直６領域分割で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域とで逆光判定が可能な分割数であれば、全画面の領域分割数を変えても、均等分割にしないでも同様の効果を得ることができる。
【００５６】
また、上記動作説明において第１領域と第２領域を図２に示した領域設定で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域であれば他の領域設定を行っても同様の効果を得ることができる。
【００５７】
また、上記動作説明において輝度信号を用いて動作説明を行ったが、ＲＧＢ信号の最大値の信号やＧ信号など輝度信号に近い信号を用いることで同様の効果を得ることができる。
【００５８】
また、上記動作説明において重み付け２全画面平均の重み係数を図３に示す重み係数で行ったが、画面上部の重み係数が小さくなっているものであれば、画面の中央の重み係数を大きくした中央重点の重み係数（例えば、図３（ｃ）（ｄ）に示す重み付け全画面平均の重み係数）でも同様の効果を得ることができる。
【００５９】
（実施の形態２）
本発明の実施の形態２について、図２ないし図７を参照しながら説明する。
【００６０】
図７は本実施の形態による撮像装置のブロック図を示す。図７において、実施の形態１による撮像装置のブロック図を示す図１と異なるのは、ピーク値２検出部７０１と、ピーク値演算部７０２が追加されたことである。すなわち、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０２の出力は、ダブルサンプリング、利得調整された後、アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調等の信号処理を行い出力端子１０６から映像信号を出力する。
【００６１】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０とピーク値２検出部７０１の入力端にも接続される。領域輝度検出部１０７では、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する。ピーク値１検出部１１０では、全画面領域内の最大輝度信号の値をピーク値１として検出する。ピーク値２検出部７０１では、画面中央を含む第２領域内の最大輝度信号の値をピーク値２として検出する。ピーク値１検出部１１０とピーク値２検出部７０１の出力端はピーク値演算部７０２の入力端に接続される。ピーク値演算部７０２では、ピーク値１とピーク値２を加重平均しピーク値を算出する。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較する。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２とピーク重み算出部１１６の入力端に接続される。平均重み算出部１１２では、平均重みを算出する。ピーク重み算出部１１６では、ピーク重みを算出する。また、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９では、それぞれ重み付け１全画面平均値と重み付け２全画面平均値を算出する。重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。全画面平均値演算部１１３では、全画面平均値を演算する。全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、平均値誤差を算出する。また、ピーク値演算部７０２の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、ピーク値誤差を算出する。ピーク値誤差算出部１１４と平均値誤差算出部１１５とピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。加重平均誤差算出部１１７では加重平均誤差を算出する。加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【００６２】
図２は本実施の形態の撮像装置における領域輝度検出部の全画面領域の領域分割を示す模式図である。図２において、２０１は第１領域、２０２は第２領域を示す。
【００６３】
図３は本実施の形態の撮像装置における重み付け１全画面平均算出部と重み付け２全画面平均算出部の重み係数を示す構成図である。図３において、（ａ）は重み付け１全画面平均値の重み係数、（ｂ）は重み付け２全画面平均値の重み係数（ｃ）は重み付け全画面平均値の重み係数の例１、（ｄ）は重み付け全画面平均値の重み係数の例２を示す。
【００６４】
図４は本実施の形態の撮像装置におけるピーク値１検出部とピーク値２検出部のピーク検出領域を示す検出領域図である。図４において、２０１は第１領域、２０２は第２領域、４０１は第１領域と第２領域をあわせた全画面領域を示す。
【００６５】
図５は本実施の形態の撮像装置における全画面平均値演算部で全画面平均値を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【００６６】
図６は本実施の形態の撮像装置における加重平均誤差算出部で加重平均誤差を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【００６７】
以上のように構成された本実施の形態の撮像装置について、以下その動作を説明する。図７において、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０３の出力は、ダブルサンプリング、利得調整された後アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調などの信号処理をして出力端子１０６から映像信号を出力する。
【００６８】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０とピーク値２検出部７０１の入力端にも接続される。
【００６９】
図２を用いて、領域輝度検出部１０７の動作説明を行う。領域輝度検出部１０７では、図２に示したように全画面領域を水平方向に８領域に分割し、垂直方向に６領域に分割する。各領域内において画素毎に輝度信号の単純加算を行い、任意の数値（例えば、２５６）で割ることによって各領域の代表輝度値を検出する。任意の数値を各領域内の画素数にすれば、代表輝度値は各領域の平均輝度値になる。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。
【００７０】
図２を用いて、第１・第２領域輝度比較部１１１の動作説明を行う。第１・第２領域輝度比較部１１１では、図２に示した画面上部を含む第１領域内の各領域の代表輝度値を平均した平均輝度から、画面中央を含む第２領域内の各領域の代表輝度値を平均した平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを算出する。輝度差分ＣＮＴが負になった場合は、負クリップを行う。また、輝度差分ＣＮＴが最大値を越えた場合は最大値クリップを行う。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２とピーク重み算出部１１６の入力端に接続される。
【００７１】
図３を用いて、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９の動作説明を行う。重み付け１全画面平均算出部１０８では、図３（ａ）に示した重み付け１全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け１全画面平均値を算出する。
【００７２】
同様に、図３（ｂ）に示した重み付け２全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け２全画面平均値を算出する。
【００７３】
図３（ｂ）に示した重み付け２全画面平均値の重み係数は画面上部になるに従い、重みが小さくなる設定になっている。画面上部に明るく輝度信号が大きい被写体がある場合は、重み付け２全画面平均値を上記の方式で算出すると、画面上部の重みが小さくなっているため、全画面の重みが均一になっている重み付け１全画面平均値より小さくなる。また逆に、画面上部に暗く輝度信号が小さい被写体がある場合は、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。
【００７４】
重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。
【００７５】
図４を用いて、ピーク値１検出部１１０とピーク値２検出部７０１の動作説明を行う。ピーク値１検出部１１０では、図４に示した全画面領域内の最大輝度信号の値をピーク値１として検出する。また、ピーク値２検出部７０１では、図４に示した画面中央を含む第2領域内の最大輝度信号の値をピーク値２として検出する。次に、ピーク値演算部７０２の動作説明を行う。ピーク値演算部７０２では、ピーク値１検出部１１０で検出されたピーク値１と、ピーク値２検出部７０１で検出されたピーク値２を加重平均しピーク値を算出する。例えば、画面上部にピーク値がある場合は、ピーク値１とピーク値２を１：１で加重平均を行いピーク値を算出すると、ピーク値はピーク値１より小さくなる。また、画面中央にピーク値がある場合は、ピーク値１とピーク値２を１：１で加重平均を行いピーク値を算出すると、ピーク値はピーク値１と等しくなる。
【００７６】
図５を用いて、平均重み算出部１１２の動作説明を行う。平均重み算出部１１２では、図５に示した第１領域と第２領域の輝度差分と平均重みの関係になるように平均重みαを算出する。平均重みαは、重み付け２全画面平均の重み係数を指し、重み付け１全画面平均の重み係数は、１００％−αで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、重み付け１全画面平均が１００％になるように平均重みαを０％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、重み付け２全画面平均が１００％になるように平均重みαを１００％にする。
【００７７】
図５を用いて、全画面平均値演算部１１３の動作説明を行う。全画面平均値演算部１１３では、重み付け１全画面平均値と重み付け２全画面平均値を平均重みαを用いて加重平均を行い、全画面平均値を算出する。全画面平均値をＡＶＥＹ、重み付け１全画面平均値をＡＶＥＹ1、重み付け２全画面平均値をＡＶＥＹ２とすると、ＡＶＥＹ＝（ＡＶＥＹ１×（１００−α）＋ＡＶＥＹ２×α）／１００、で示される式によって、全画面平均値を算出することができる。
【００７８】
全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、所定の目標値から全画面平均値を減算して平均値誤差を算出する。
【００７９】
また、ピーク値演算部７０２の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、所定の目標値からピーク値を減算してピーク値誤差を算出する。
【００８０】
ここで、絞り１０２の開閉動作が安定した状態では通常、ピーク値誤差は負の値になり、平均値誤差は正の値になる。ピーク値誤差と平均値誤差を加重平均し打ち消しあって加重平均誤差が０になる絞り位置で絞り１０２が釣り合うようになっている。ピーク値誤差算出部１１４と平均値誤差算出部１１５の出力端は加重平均誤差算出部１１７の入力端に接続される。
【００８１】
図６を用いて、ピーク重み算出部１１６の動作説明を行う。ピーク重み算出部１１６では、図６に示した第１領域と第２領域の輝度差分とピーク重みの関係になるようにピーク重みβを算出する。ピーク重みβは、ピーク値誤差の重み係数を指し、平均値誤差の重み係数は、１００％−βで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、ピーク値誤差が２５％になるようにピーク重みβを２５％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、平均値誤差が１００％になるようにピーク重みβを０％にする。ピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。
【００８２】
図６を用いて、加重平均誤差算出部１１７の動作説明を行う。加重平均誤差算出部１１７では、ピーク値誤差算出部１１４で得られたピーク値誤差と平均値誤差算出部１１５で得られた平均値誤差をピーク重み算出部１１６で得られたピーク重みβを用いて加重平均を行い、加重平均誤差を算出する。加重平均誤差をＧＯＳＡ、ピーク値誤差をＰＧＯＳＡ、平均値誤差をＡＧＯＳＡとすると、ＧＯＳＡ＝（ＡＧＯＳＡ×（１００−β）＋ＰＧＯＳＡ×β）／１００、で示される式で加重平均誤差を算出することができる。
【００８３】
加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【００８４】
逆光判定の動作について、図２ないし図７を用いて動作説明を行う。逆光判定は、第１・第２領域輝度比較部１１１で行う。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度から、画面中央を含む第２領域の平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを演算する。逆光時に人などの主被写体を画面中央にして撮影すると、主被写体は暗くなり、画面上部には空や雲など明るい被写体になることが多いので、第1領域と第２領域の輝度差分ＣＮＴは正の値になる。強逆光になるほど、第1領域と第２領域の輝度差分ＣＮＴの値は大きくなる。すなわち、第1領域と第２領域の輝度差分ＣＮＴが正になるか負になるかで逆光判定を行い、輝度差分ＣＮＴの値の大きさで逆光の強度を判断する。また、過順光時に主被写体を画面中央にして撮影すると、主被写体が明るくなっているので、第1領域と第２領域の輝度差分ＣＮＴは負の値になる。
【００８５】
次に、逆光判定された時の逆光補正の動作について、図２ないし図７を用いて動作説明を行う。逆光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが正の値になるので、第２領域に比べ第１領域に明るい被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より小さくなる。
【００８６】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが大きな値になっていれば、平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け２全画面平均に近い全画面平均値を算出する。よって、全画面平均値は従来の撮像装置の全画面平均値より小さくなる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の全画面平均値より小さくなっているので、平均値誤差は大きくなる。
【００８７】
また、逆光時は第２領域に比べ第１領域に明るい被写体が多いので、全画面領域のピーク値１は、画面中央を含む第２領域のピーク値２より大きくなる。ここで、ピーク値演算部７０２で、ピーク値１とピーク値２を１：１で加重平均しピーク値を算出すると、ピーク値は、従来の撮像装置が全画面領域で検出したピーク値より小さくなる。次に、ピーク値誤差算出部１１４で所定の目標値からピーク値を減算してピーク値誤差を算出する。ピーク値は従来のピーク値より小さくなっているので、ピーク値誤差の絶対値は小さくなる。
【００８８】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。従って、逆光時は、ピーク重みβは小さくなる。よって、加重平均誤差算出部１１７で算出される加重平均誤差は平均値誤差に近い値になる。平均値誤差は大きくなり、ピーク値誤差は小さくなっているので加重平均誤差も大きくなる。加重平均誤差が大きいと、加重平均誤差を小さくするように、絞り１０２を開く方向に絞り制御部１１８が働くので、逆光時は従来の撮像装置より絞り１０２を開き逆光補正を行うことができる。
【００８９】
次に、順光・過順光判定された時の絞り制御の動作について、図２ないし図７を用いて動作説明を行う。順光・過順光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になるので、画面中央を含む第２領域に比べ画面上部を含む第１領域に暗い被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。また、重み付け１全画面平均値の重み係数は均等なっているので、重み付け１全画面平均値は従来の撮像装置の全画面平均値と等しくなる。
【００９０】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になっていれば、負クリップされる。平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが０の時は、重み付け１全画面平均の重みが１００％になるように平均重みを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け１全画面平均を全画面平均値として算出する。よって、全画面平均値は従来の撮像装置の全画面平均値と同じになる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の撮像装置の全画面平均値と同じなので、平均値誤差も等しくなる。
【００９１】
また、順光・過順光時は第１領域に比べ第２領域に明るい被写体が多いので、全画面領域のピーク値１と、画面中央を含む第２領域のピーク値２は等しくなる。ここで、ピーク値演算部７０２で、ピーク値１とピーク値２を１：１で加重平均しピーク値を算出すると、ピーク値は、従来の撮像装置が全画面領域で検出したピーク値と等しくなる。次に、ピーク値誤差算出部１１４で所定の目標値からピーク値を減算してピーク値誤差を算出する。ピーク値は従来のピーク値と等しくなっているので、ピーク値誤差も等しくなる。
【００９２】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが負クリップされた０になるので、ピークの重み係数が最大値になるようにピーク重みを算出する。従って、順光・過順光時は、ピーク重みは２５％になる。よって、加重平均誤差算出部１１７では平均値誤差とピーク値誤差を３：１で加重平均を行い加重平均誤差を算出する。
【００９３】
ピーク重みの最大値を従来撮像装置の設定と等しく設定すれば、平均値誤差とピーク値誤差が従来の撮像装置と等しいので、加重平均誤差も従来の撮像装置と等しくなる。よって、順光・過順光時は従来の撮像装置の絞り制御と同等の制御を行うことができる。
【００９４】
従って、逆光時は従来の撮像装置より絞りを開くことができ、順光・過順光時は従来の撮像装置と同等の絞り制御になるので、適切な露光制御を行うことができる撮像装置を得ることができる。
【００９５】
このように構成することによって、第１・第２領域輝度比較部１１１で画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較することで逆光判定を行う。逆光判定時は、平均値誤差の重み係数がピーク値誤差の重み係数より大きくなるように加重平均のピーク重みを算出し、更に重み付け１全画面平均値の重み係数が重み付け２全画面平均値の重み係数より小さくなるように加重平均の平均重みを算出することで、絞りを更に開くことができる。更にピーク１輝度値とピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができる。また、順光判定時は、従来の撮像装置と同等の絞り制御を行うことができるので、順光被写体や強逆光被写体に対しても、逆光、順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果を得られる。
【００９６】
なお、上記動作説明において全画面の領域分割を水平８領域分割、垂直６領域分割で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域とで逆光判定が可能な分割数であれば、全画面の領域分割数を変えても、均等分割にしないでも同様の効果を得ることができる。
【００９７】
また、上記動作説明において第１領域と第２領域を図２に示した領域設定で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域であれば他の領域設定を行っても同様の効果を得ることができる。
【００９８】
また、上記動作説明において輝度信号を用いて動作説明を行ったが、ＲＧＢ信号の最大値の信号やＧ信号など輝度信号に近い信号を用いることで同様の効果を得ることができる。
【００９９】
また、上記動作説明において重み付け２全画面平均の重み係数を図３に示す重み係数で行ったが、画面上部の重み係数が小さくなっているものであれば、画面の中央の重み係数を大きくした中央重点の重み係数（例えば、図３（ｃ）（ｄ）に示す重み付け全画面平均の重み係数）でも同様の効果を得ることができる。
【０１００】
（実施の形態３）
本発明の実施の形態３について、図２ないし図４と図６、図８を参照しながら説明する。
【０１０１】
図８は本実施の形態による撮像装置のブロック図を示す。図８において、実施の形態２による撮像装置のブロック図を示す図７と異なるのは、重み付け１全画面平均算出部１０８と、平均重み演算部１１２と、全画面平均値演算部１１３が削除されたことである。すなわち、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０２の出力は、ダブルサンプリング、利得調整された後アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調等の信号処理を行い出力端子１０６から映像信号を出力する。
【０１０２】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０とピーク値２検出部７０１の入力端にも接続される。領域輝度検出部１０７では、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する。ピーク値１検出部１１０では、全画面領域内の最大輝度信号の値をピーク値１として検出する。ピーク値２検出部７０１では、画面中央を含む第２領域内の最大輝度信号の値をピーク値２として検出する。ピーク値１検出部１１０とピーク値２検出部７０１の出力端はピーク値演算部７０２の入力端に接続される。ピーク値演算部７０２では、ピーク値１とピーク値２を加重平均しピーク値を算出する。領域輝度検出部１０７の出力端は、重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較する。第１・第２領域輝度比較部１１１の出力端は、ピーク重み算出部１１６の入力端に接続される。ピーク重み算出部１１６では、ピーク重みを算出する。また、重み付け２全画面平均算出部１０９では、重み付け２全画面平均値を算出する。重み付け２全画面平均算出部１０９の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、平均値誤差を算出する。また、ピーク値演算部７０２の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、ピーク値誤差を算出する。ピーク値誤差算出部１１４と平均値誤差算出部１１５とピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。加重平均誤差算出部１１７では加重平均誤差を算出する。加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【０１０３】
図２は本実施の形態の撮像装置における領域輝度検出部の全画面領域の領域分割を示す模式図である。図２において、２０１は第１領域、２０２は第２領域を示す。
【０１０４】
図３は本実施の形態の撮像装置における重み付け１全画面平均算出部と重み付け２全画面平均算出部の重み係数を示す構成図である。図３において、（ａ）は重み付け１全画面平均値の重み係数、（ｂ）は重み付け２全画面平均値の重み係数（ｃ）は重み付け全画面平均値の重み係数の例１、（ｄ）は重み付け全画面平均値の重み係数の例２を示す。
【０１０５】
図４は本実施の形態の撮像装置におけるピーク値１検出部とピーク値２検出部のピーク検出領域を示す検出領域図である。図４において、２０１は第１領域、２０２は第２領域、４０１は第１領域と第２領域をあわせた全画面領域を示す。
【０１０６】
図６は本実施の形態の撮像装置における加重平均誤差算出部で加重平均誤差を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【０１０７】
以上のように構成された本実施の形態の撮像装置について、以下その動作を説明する。図８において、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０３の出力は、ダブルサンプリング、利得調整された後、アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調などの信号処理をして出力端子１０６から映像信号を出力する。
【０１０８】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７とピーク値１検出部１１０とピーク値２検出部７０１の入力端にも接続される。
【０１０９】
図２を用いて、領域輝度検出部１０７の動作説明を行う。領域輝度検出部１０７では、図２に示したように全画面領域を水平方向に８領域に分割し、垂直方向に６領域に分割する。各領域内において画素毎に輝度信号の単純加算を行い、任意の数値（例えば、２５６）で割ることによって各領域の代表輝度値を検出する。任意の数値を各領域内の画素数にすれば、代表輝度値は各領域の平均輝度値になる。領域輝度検出部１０７の出力端は、重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。
【０１１０】
図２を用いて、第１・第２領域輝度比較部１１１の動作説明を行う。第１・第２領域輝度比較部１１１では、図２に示した画面上部を含む第１領域内の各領域の代表輝度値を平均した平均輝度から、画面中央を含む第２領域内の各領域の代表輝度値を平均した平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを算出する。輝度差分ＣＮＴが負になった場合は、負クリップを行う。また、輝度差分ＣＮＴが最大値を越えた場合は最大値クリップを行う。第１・第２領域輝度比較部１１１の出力端は、ピーク重み算出部１１６の入力端に接続される。
【０１１１】
図３を用いて、重み付け２全画面平均算出部１０９の動作説明を行う。重み付け２全画面平均算出部１０９では、図３（ｂ）に示した重み付け２全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け２全画面平均値を算出する。この重み付け２全画面平均値を全画面平均値として出力する。
【０１１２】
図３（ｂ）に示した重み付け２全画面平均値の重み係数は画面上部になるに従い、重みが小さくなる設定になっている。図３（ｂ）に示した重み付け２全画面平均値の重み係数を用いて重み付け２全画面平均値を算出すると、画面上部に明るく輝度信号が大きい被写体の時、画面上部の重みが小さくなっているため、全画面の重みが均一になっている重み付け１全画面平均値より小さくなる。また逆に、画面上部に暗く輝度信号が小さい被写体がある場合は、重み付け１全画面平均値より大きくなる。
【０１１３】
図４を用いて、ピーク値１検出部１１０とピーク値２検出部７０１の動作説明を行う。ピーク値１検出部１１０では、図４に示した全画面領域内の最大輝度信号の値をピーク値１として検出する。また、ピーク値２検出部７０１では、図４に示した画面中央を含む第2領域内の最大輝度信号の値をピーク値２として検出する。次に、ピーク値演算部７０２の動作説明を行う。ピーク値演算部７０２では、ピーク値１検出部１１０で検出されたピーク値１と、ピーク値２検出部７０１で検出されたピーク値２を加重平均しピーク値を算出する。例えば、画面上部にピーク値がある場合は、ピーク値１とピーク値２を１：１で加重平均を行いピーク値を算出すると、ピーク値はピーク値１より小さくなる。また、画面中央にピーク値がある場合は、ピーク値１とピーク値２を１：１で加重平均を行いピーク値を算出すると、ピーク値はピーク値１と等しくなる。
【０１１４】
重み付け２全画面平均算出部１０９の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、所定の目標値から全画面平均値を減算して平均値誤差を算出する。
【０１１５】
また、ピーク値演算部７０２の出力端はピーク値誤差算出部１１４の入力端に接続される。ピーク値誤差算出部１１４では、所定の目標値からピーク値を減算してピーク値誤差を算出する。
【０１１６】
ここで、絞り１０２の開閉動作が安定した状態では通常、ピーク値誤差は負の値になり、平均値誤差は正の値になる。ピーク値誤差と平均値誤差を加重平均し打ち消しあって加重平均誤差が０になる絞り位置で絞り１０２が釣り合うようになっている。ピーク値誤差算出部１１４と平均値誤差算出部１１５の出力端は加重平均誤差算出部１１７の入力端に接続される。
【０１１７】
図６を用いて、ピーク重み算出部１１６の動作説明を行う。ピーク重み算出部１１６では、図６に示した第１領域と第２領域の輝度差分とピーク重みの関係になるようにピーク重みβを算出する。ピーク重みβは、ピーク値誤差の重み係数を指し、平均値誤差の重み係数は、１００％−βで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、ピーク値誤差が２５％になるようにピーク重みβを２５％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、平均値誤差が１００％になるようにピーク重みβを０％にする。ピーク重み算出部１１６の出力端は加重平均誤差算出部１１７の入力端に接続される。
【０１１８】
図６を用いて、加重平均誤差算出部１１７の動作説明を行う。加重平均誤差算出部１１７では、ピーク値誤差算出部１１４で得られたピーク値誤差と平均値誤差算出部１１５で得られた平均値誤差をピーク重み算出部１１６で得られたピーク重みβを用いて加重平均を行い、加重平均誤差を算出する。加重平均誤差をＧＯＳＡ、ピーク値誤差をＰＧＯＳＡ、平均値誤差をＡＧＯＳＡとすると、ＧＯＳＡ＝（ＡＧＯＳＡ×（１００−β）＋ＰＧＯＳＡ×β）／１００、で示される式で加重平均誤差を算出することができる。
【０１１９】
加重平均誤差算出部１１７の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【０１２０】
逆光判定の動作について、図２ないし図４と図８を用いて動作説明を行う。逆光判定は、第１・第２領域輝度比較部１１１で行う。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度から、画面中央を含む第２領域の平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを演算する。逆光時に人などの主被写体を画面中央にして撮影すると、主被写体は暗くなり、画面上部には空や雲など明るい被写体になることが多いので、第1領域と第２領域の輝度差分ＣＮＴは正の値になる。強逆光になるほど、第1領域と第２領域の輝度差分ＣＮＴの値は大きくなる。すなわち、第1領域と第２領域の輝度差分ＣＮＴが正になるか負になるかで逆光判定を行い、輝度差分ＣＮＴの値の大きさで逆光の強度を判断する。また、過順光時に主被写体を画面中央にして撮影すると、主被写体が明るくなっているので、第1領域と第２領域の輝度差分ＣＮＴは負の値になる。
【０１２１】
次に、逆光判定された時の逆光補正の動作について、図２ないし図４と図６、図８を用いて動作説明を行う。逆光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが正の値になるので、第２領域に比べ第１領域に明るい被写体が多いことになる。この時、重み付け２全画面平均算出部１０９で図３（ｂ）に示す重み付け２全画面平均値の重み係数を用いて重み付け２全画面平均値を算出すると、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値は従来の撮像装置の全画面平均値より小さくなる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の全画面平均値より小さくなっているので、平均値誤差は大きくなる。
【０１２２】
また、逆光時は第２領域に比べ第１領域に明るい被写体が多いので、全画面領域のピーク値１は、画面中央を含む第２領域のピーク値２より大きくなる。ここで、ピーク値演算部７０２で、ピーク値１とピーク値２を１：１で加重平均しピーク値を算出すると、ピーク値は、従来の全画面領域で検出したピーク値より小さくなる。次に、ピーク値誤差算出部１１４で所定の目標値からピーク値を減算してピーク値誤差を算出する。ピーク値は従来のピーク値より小さくなっているので、ピーク値誤差の絶対値は小さくなる。
【０１２３】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、ピークの重み係数が小さくなるようにピーク重みβを算出する。従って、逆光時は、ピーク重みβは小さくなる。よって、加重平均誤差算出部１１７で算出される加重平均誤差は平均値誤差に近い値になる。平均値誤差は大きくなり、ピーク値誤差は小さくなっているので加重平均誤差も大きくなる。加重平均誤差が大きいと、加重平均誤差を小さくするように、絞り１０２を開く方向に絞り制御部１１８が働くので、逆光時は従来の撮像装置より絞り１０２を開き逆光補正を行うことができる。
【０１２４】
次に、過順光判定された時の絞り制御の動作について、図２ないし図４と図６、図８を用いて動作説明を行う。順光・過順光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になるので、画面中央を含む第２領域に比べ画面上部を含む第１領域に暗い被写体が多いことになる。この時、重み付け２全画面平均算出部１０９で、図３（ｂ）に示す重み付け２全画面平均値の重み係数を用いて重み付け２全画面平均値を算出すると、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値は従来の撮像装置の全画面平均値より大きくなる。全画面平均値は従来の撮像装置の全画面平均値より大きくなっているので、平均値誤差は小さくなる。
【０１２５】
また、過順光時は第１領域に比べ第２領域に明るい被写体が多いので、全画面領域のピーク値１と、画面中央を含む第２領域のピーク値２は等しくなる。ここで、ピーク値演算部７０２で、ピーク値１とピーク値２を１：１で加重平均しピーク値を算出すると、ピーク値は、従来の撮像装置が全画面領域で検出したピーク値と等しくなる。次に、ピーク値誤差算出部１１４で所定の目標値からピーク値を減算してピーク値誤差を算出する。ピーク値は従来の撮像装置のピーク値と等しくなっているので、ピーク値誤差も等しくなる。
【０１２６】
また、ピーク重み算出部１１６は、第１領域と第２領域の輝度差分ＣＮＴが負クリップされた０になるので、ピークの重み係数が最大値になるようにピーク重みを算出する。従って、順光・過順光時は、ピーク重みは２５％になる。よって、加重平均誤差算出部１１７では平均値誤差とピーク値誤差を３：１で加重平均を行い加重平均誤差を算出する。
【０１２７】
ピーク重みの最大値を従来撮像装置の設定と等しく設定すれば、従来の撮像装置より平均値誤差は小さくなり、ピーク値誤差は等しいので、加重平均誤差は従来の撮像装置よりも小さくなる。よって、過順光時は従来の撮像装置より絞りを閉じる絞り制御を行うことができる。
【０１２８】
従って、逆光時は従来の撮像装置より絞りを開くことができ、過順光時は従来の撮像装置より絞りを閉じる絞り制御になるので、逆光時は絞りが開き足りない過順光時は絞りの開きすぎを抑えることができ適切な露光制御を行うことができる撮像装置を得ることができる。
【０１２９】
このように構成することによって、第１・第２領域輝度比較部１１１で画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較することで逆光判定を行う。逆光判定時は、平均値誤差の重み係数がピーク値誤差の重み係数より大きくなるように加重平均のピーク重みを算出することで、絞りを開くことができる。更にピーク１輝度値とピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができる。また、過順光判定時は、絞りの開きすぎを抑えることができるので、過順光被写体や強逆光被写体に対しても、逆光、順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果を得られる。
【０１３０】
なお、上記動作説明において全画面の領域分割を水平８領域分割、垂直６領域分割で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域とで逆光判定が可能な分割数であれば、全画面の領域分割数を変えても、均等分割にしないでも同様の効果を得ることができる。
【０１３１】
また、上記動作説明において第１領域と第２領域を図２に示した領域設定で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域であれば他の領域設定を行っても同様の効果を得ることができる。
【０１３２】
また、上記動作説明において輝度信号を用いて動作説明を行ったが、ＲＧＢ信号の最大値の信号やＧ信号など輝度信号に近い信号を用いることで同様の効果を得ることができる。
【０１３３】
また、上記動作説明において重み付け２全画面平均の重み係数を図３に示す重み係数で行ったが、画面上部の重み係数が小さくなっているものであれば、画面の中央の重み係数を大きくした中央重点の重み係数（例えば、図３（ｃ）（ｄ）に示す重み付け全画面平均の重み係数）でも同様の効果を得ることができる。
【０１３４】
（実施の形態４）
本発明の実施の形態４について、図２、図３、図５と図９を参照しながら説明する。
【０１３５】
図９は本実施の形態による撮像装置のブロック図を示す。図９において、実施の形態１による撮像装置のブロック図を示す図１と異なるのは、ピーク値１検出部１１０と、ピーク値誤差算出部１１４と、ピーク重み算出部１１６と、加重平均誤差算出部１１７を削除したところである。すなわち、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０２の出力は、ダブルサンプリング、利得調整された後アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調等の信号処理を行い出力端子１０６から映像信号を出力する。
【０１３６】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７の入力端に接続される。領域輝度検出部１０７では、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較する。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２の入力端に接続される。平均重み算出部１１２では、平均重みを算出する。また、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９では、それぞれ重み付け１全画面平均値と重み付け２全画面平均値を算出する。重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。全画面平均値演算部１１３では、全画面平均値を演算する。全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、平均値誤差を算出する。平均値誤差算出部１１５の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【０１３７】
図２は本実施の形態の撮像装置における領域輝度検出部の全画面領域の領域分割を示す模式図である。図２において、２０１は第１領域、２０２は第２領域を示す。
【０１３８】
図３は本実施の形態の撮像装置における重み付け１全画面平均算出部と重み付け２全画面平均算出部の重み係数を示す構成図である。図３において、（ａ）は重み付け１全画面平均値の重み係数、（ｂ）は重み付け２全画面平均値の重み係数（ｃ）は重み付け全画面平均値の重み係数の例１、（ｄ）は重み付け全画面平均値の重み係数の例２を示す。
【０１３９】
図５は本実施の形態の撮像装置における全画面平均値演算部で全画面平均値を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図である。
【０１４０】
以上のように構成された本実施の形態の撮像装置について、以下その動作を説明する。図９において、光学系１０１により形成された被写体の光学像は絞り１０２を経て撮像素子(ＣＣＤ)１０３に入射され光電変換される。撮像素子１０３の出力端はＡＳＰ・Ａ／Ｄ変換器１０４の入力端に接続され、撮像素子１０３の出力は、ダブルサンプリング、利得調整された後アナログ／デジタル変換される。ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は信号処理回路１０５の入力端に接続される。信号処理回路１０５では、輪郭強調などの信号処理をして出力端子１０６から映像信号を出力する。
【０１４１】
また、ＡＳＰ・Ａ／Ｄ変換器１０４の出力端は、領域輝度検出部１０７の入力端にも接続される。
【０１４２】
図２を用いて、領域輝度検出部１０７の動作説明を行う。領域輝度検出部１０７では、図２に示したように全画面領域を水平方向に８領域に分割し、垂直方向に６領域に分割する。各領域内において画素毎に輝度信号の単純加算を行い、任意の数値（例えば、２５６）で割ることによって各領域の代表輝度値を検出する。任意の数値を各領域内の画素数にすれば、代表輝度値は各領域の平均輝度値になる。領域輝度検出部１０７の出力端は、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と第１・第２領域輝度比較部１１１の入力端に接続される。
【０１４３】
図２を用いて、第１・第２領域輝度比較部１１１の動作説明を行う。第１・第２領域輝度比較部１１１では、図２に示した画面上部を含む第１領域内の各領域の代表輝度値を平均した平均輝度から、画面中央を含む第２領域内の各領域の代表輝度値を平均した平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを演算する。輝度差分ＣＮＴが負になった場合は、負クリップを行う。また、輝度差分ＣＮＴが最大値を越えた場合は最大値クリップを行う。第１・第２領域輝度比較部１１１の出力端は、平均重み算出部１１２の入力端に接続される。
【０１４４】
図３を用いて、重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９の動作説明を行う。重み付け１全画面平均算出部１０８では、図３（ａ）に示した重み付け１全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け１全画面平均値を算出する。
【０１４５】
同様に、図３（ｂ）に示した重み付け２全画面平均値の重み係数を各領域の代表輝度にそれぞれ乗算し、各領域の乗算結果を全て加算する。次に各領域の重み係数を全て加算する。最後に、乗算結果の全加算値を重み係数の全加算値で割ることによって重み付け２全画面平均値を算出する。
【０１４６】
図３（ｂ）に示した重み付け２全画面平均値の重み係数は画面上部になるに従い、重みが小さくなる設定になっている。画面上部に明るく輝度信号が大きい被写体がある場合、重み付け２全画面平均値を上記の方式で算出すると、画面上部の重みが小さくなっているため、全画面の重みが均一になっている重み付け１全画面平均値より小さくなる。また逆に、画面上部に暗く輝度信号が小さい被写体がある場合は、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。
【０１４７】
重み付け１全画面平均算出部１０８と重み付け２全画面平均算出部１０９と平均重み算出部１１２の出力端は全画面平均値演算部１１３の入力端に接続される。
【０１４８】
図５を用いて、平均重み算出部１１２の動作説明を行う。平均重み算出部１１２では、図５に示した第１領域と第２領域の輝度差分と平均重みの関係になるように平均重みαを算出する。平均重みαは、重み付け２全画面平均の重み係数を指し、重み付け１全画面平均の重み係数は、１００％−αで求める。すなわち、第１領域と第２領域の輝度差分ＣＮＴが負の時は、重み付け１全画面平均が１００％になるように平均重みαを０％にする。第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。第１領域と第２領域の輝度差分ＣＮＴが最大値になった場合、重み付け２全画面平均が１００％になるように平均重みαを１００％にする。
【０１４９】
図５を用いて、全画面平均値演算部１１３の動作説明を行う。全画面平均値演算部１１３では、重み付け１全画面平均値と重み付け２全画面平均値を平均重みαを用いて加重平均を行い、全画面平均値を算出する。全画面平均値をＡＶＥＹ、重み付け１全画面平均値をＡＶＥＹ1、重み付け２全画面平均値をＡＶＥＹ２とすると、ＡＶＥＹ＝（ＡＶＥＹ１×（１００−α）＋ＡＶＥＹ２×α）／１００、で示す式によって、全画面平均値を算出することができる。
【０１５０】
全画面平均値演算部１１３の出力端は平均値誤差算出部１１５の入力端に接続される。平均値誤差算出部１１５では、所定の目標値から全画面平均値を減算して平均値誤差を算出する。
【０１５１】
平均値誤差算出部１１５の出力端は絞り制御部１１８の入力端に接続される。絞り制御部１１８では、加重平均誤差を０へ追従させるべく絞り開度を決定する。
【０１５２】
逆光判定の動作について、図２と図９を用いて動作説明を行う。逆光判定は、第１・第２領域輝度比較部１１１で行う。第１・第２領域輝度比較部１１１では、画面上部を含む第１領域の平均輝度から、画面中央を含む第２領域の平均輝度を減算し、第1領域と第２領域の輝度差分ＣＮＴを演算する。逆光時に人などの主被写体を画面中央にして撮影すると、主被写体は暗くなり、画面上部には空や雲など明るい被写体になることが多いので、第1領域と第２領域の輝度差分ＣＮＴは正の値になる。強逆光になるほど、第1領域と第２領域の輝度差分ＣＮＴの値は大きくなる。すなわち、第1領域と第２領域の輝度差分ＣＮＴが正になるか負になるかで逆光判定を行い、輝度差分ＣＮＴの値の大きさで逆光の強度を判断する。また、過順光時に主被写体を画面中央にして撮影すると、主被写体が明るくなっているので、第1領域と第２領域の輝度差分ＣＮＴは負の値になる。
【０１５３】
次に、逆光判定された時の逆光補正の動作について、図２、図３、図５と図９を用いて動作説明を行う。逆光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが正の値になるので、第２領域に比べ第１領域に明るい被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より小さくなる。
【０１５４】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが大きな値になっていれば、平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが徐々に大きくなるに従って、重み付け２全画面平均の重みが大きくなるように平均重みαを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け２全画面平均に近い全画面平均値を算出する。よって、全画面平均値は従来の全画面平均値より小さくなる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の全画面平均値より小さくなっているので、平均値誤差は大きくなる。
【０１５５】
平均値誤差が大きいと、平均値誤差を小さくするように、絞り１０２を開く方向に絞り制御部１１８が働くので、逆光時は従来の撮像装置より絞り１０２を開き逆光補正を行うことができる。
【０１５６】
次に、順光・過順光判定された時の絞り制御の動作について、図２、図３、図５と図９を用いて動作説明を行う。順光・過順光判定時は、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になるので、画面中央を含む第２領域に比べ画面上部を含む第１領域に暗い被写体が多いことになる。この時、重み付け１全画面平均値と重み付け２全画面平均値を算出すると、重み付け１全画面平均値の重み係数は均等なっており、重み付け２全画面平均値の重み係数は画面上部になるほど重み係数が小さくなっているので、重み付け２全画面平均値が重み付け１全画面平均値より大きくなる。また、重み付け１全画面平均値の重み係数は均等なっているので、重み付け１全画面平均値は従来の撮像装置の全画面平均値と等しくなる。
【０１５７】
また、第１・第２領域輝度比較部１１１で算出した第1領域と第２領域の輝度差分ＣＮＴが負の値になっていれば、負クリップされる。平均重み算出部１１２は第１領域と第２領域の輝度差分ＣＮＴが０の時は、重み付け１全画面平均の重みが１００％になるように平均重みを算出する。従って、全画面平均値演算部１１３は、加重平均によって重み付け１全画面平均を全画面平均値として算出する。よって、全画面平均値は従来の撮像装置の全画面平均値と同じになる。次に、平均値誤差算出部１１５で所定の目標値から全画面平均値を減算して平均値誤差を算出する。全画面平均値は従来の撮像装置の全画面平均値と同じなので、平均値誤差も等しくなる。よって、順光・過順光時は従来の撮像装置の絞り制御と同等の制御を行うことができる。
【０１５８】
従って、逆光時は従来の撮像装置より絞りを開くことができ、順光・過順光時は従来の撮像装置と同等の絞り制御になるので、適切な露光制御を行うことができる撮像装置を得ることができる。
【０１５９】
このように構成することによって、第１・第２領域輝度比較部１１１で画面上部を含む第１領域の平均輝度と、画面中央を含む第２領域の平均輝度を比較することで逆光判定を行う。逆光判定時は、重み付け１全画面平均値の重み係数が重み付け２全画面平均値の重み係数より小さくなるように加重平均の平均重みを算出することで、絞りを更に開くことができる。また、順光判定時は、従来の撮像装置と同等の絞り制御を行うことができるので、順光被写体や強逆光被写体に対しても、逆光、順光の判定を行い、適切な露光制御を行うことができる撮像装置を提供することができるという効果を得られる。
【０１６０】
なお、上記動作説明において全画面の領域分割を水平８領域分割、垂直６領域分割で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域とで逆光判定が可能な分割数であれば、全画面の領域分割数を変えても、均等分割にしないでも同様の効果を得ることができる。
【０１６１】
また、上記動作説明において第１領域と第２領域を図２に示した領域設定で動作説明を行ったが、画面上部を含む第１領域と、画面中央を含む第２領域であれば他の領域設定を行っても同様の効果を得ることができる。
【０１６２】
また、上記動作説明において輝度信号を用いて動作説明を行ったが、ＲＧＢ信号の最大値の信号やＧ信号など輝度信号に近い信号を用いることで同様の効果を得ることができる。
【０１６３】
また、上記動作説明において重み付け２全画面平均の重み係数を図３に示す重み係数で行ったが、画面上部の重み係数が小さくなっているものであれば、画面の中央の重み係数を大きくした中央重点の重み係数（例えば、図３（ｃ）（ｄ）に示す重み付け全画面平均の重み係数）でも同様の効果を得ることができる。
【０１６４】
【発明の効果】
以上のように、本発明の第１の発明の撮像装置によれば、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、全画面領域内のピーク輝度値を検出するピーク値１検出部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値１検出部で得られたピーク輝度値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備え、前記第１・第２領域輝度比較部において画面上部を含む前記第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正であるか負であるかによって逆光判定を行う。この演算結果によって、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出し、更に、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出することで、絞りを更に開くことができるように構成したことにより、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、逆光判定時は従来の撮像装置より絞りを開く逆光補正を行い、順光判定時は従来の撮像装置と同じ絞り制御を行うことで、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【０１６５】
また、本発明の第２の発明の撮像装置によれば、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、全画面領域内のピーク１輝度値を検出するピーク値１検出部と、画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備え、前記第１・第２領域輝度比較部において画面上部を含む前記第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の前記輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正であるか負であるかによって逆光判定を行う。この演算結果によって、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出し、更に前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出することで、絞りを更に開くことができる。更にピーク１輝度値とピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができるように構成したことにより、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、逆光判定時は従来の撮像装置より絞りを開く逆光補正を行い、順光判定時は従来の撮像装置と同じ絞り制御を行うことで、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【０１６６】
また、本発明の第３の発明の撮像装置によれば、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、全画面領域内のピーク１輝度値を検出するピーク値１検出部と、画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、
前記重み付け２全画面平均算出部で得られた重み付け２全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記全画面平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備え、前記第１・第２領域輝度比較部において画面上部を含む前記第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の前記輝度差分ＣＮＴの値を演算し、前記輝度差分ＣＮＴの値が正であるか負であるかによって逆光判定を行う。この演算結果によって、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出することで、絞りを開くことができ、更に前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出することで、ピーク値が画面上部だけにある時はピーク値を小さくすることができ、絞りを更に開くことができるように構成したことにより、順光被写体や強逆光被写体に対しても、逆光、過過順光の判定を行い、逆光判定時は従来の撮像装置より絞りを開く逆光補正を行い、過順光判定時は従来の撮像装置より少し絞りを閉じる制御を行うことで、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【０１６７】
また、本発明の第４の発明の撮像装置によれば、全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、前記平均値誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、輝度差分ＣＮＴの値が正の場合は逆光と判断し、輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、前記第１・第２領域輝度比較部の演算結果から、輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部とを備え、前記第１・第２領域輝度比較部において画面上部を含む前記第１の領域の平均輝度から画面中央を含む前記第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の輝度差分ＣＮＴの値を演算し、この輝度差分ＣＮＴの値が正であるか負であるかによって逆光判定を行う。この演算結果によって、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出することで、絞りを更に開くことができるように構成したことにより、順光被写体や強逆光被写体に対しても、逆光、過順光の判定を行い、逆光判定時は従来の撮像装置より絞りを開く逆光補正を行い、順光判定時は従来の撮像装置と同じ絞り制御を行うことで、適切な露光制御を行うことができる撮像装置を提供することができるという効果が得られる。
【図面の簡単な説明】
【図１】本発明の実施の形態１による撮像装置を示すブロック図
【図２】同撮像装置の領域輝度検出部での全画面領域の領域分割を示す模式図
【図３】同撮像装置の重み付け１全画面平均算出部と重み付け２全画面平均算出部の重み係数を示す構成図
【図４】同撮像装置のピーク値１検出部とピーク値２検出部のピーク検出領域を示す検出領域図
【図５】同撮像装置の全画面平均値演算部で全画面平均値を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図
【図６】同撮像装置の加重平均誤差算出部で加重平均誤差を算出する時の第１領域と第２領域の輝度差分と重みの関係を示した関係図
【図７】本発明の実施の形態２による撮像装置を示すブロック図
【図８】本発明の実施の形態３による撮像装置を示すブロック図
【図９】本発明の実施の形態４による撮像装置を示すブロック図
【図１０】従来の撮像装置を示すブロック図
【符号の説明】
１０１ 光学系
１０２ 絞り
１０３ 撮像素子
１０４ ＡＳＰ・Ａ／Ｄ変換器
１０５ 信号処理回路
１０７ 領域輝度検出部
１０８ 重み付け１全画面平均算出部
１０９ 重み付け２全画面平均算出部
１１０ ピーク値１検出部
１１１ 第１・第２領域輝度比較部
１１２ 平均重み算出部
１１３ 全画面平均値演算部
１１４ ピーク値誤差演算部
１１５ 平均値誤差演算部
１１６ ピーク重み算出部
１１７ 加重平均誤差算出部
１１８ 絞り制御部
７０１ ピーク値２検出部
７０２ ピーク値演算部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an imaging apparatus capable of performing appropriate exposure control by performing backlight determination and continuously switching control based on a weighted average value according to a determination result.
[0002]
[Prior art]
A conventional imaging apparatus is known as described in JP-A-6-31423.
[0003]
FIG. 10 is a block diagram of a conventional imaging apparatus disclosed in the publication. In FIG. 10, 1001 is an optical system, 1002 is an aperture, 1003 is an image sensor, 1004 is an amplifier, 1005 is a signal processing circuit, 1006 is an output terminal, 1007 is an all-screen average detection unit, 1008 is an aperture control unit, and 1009 is a first control unit. 1 and 2 region luminance comparison unit, 1010 is a peak value detection unit, 1011 is a peak value error calculation unit, 1012 is an average value error calculation unit, 1013 is a weighted average error calculation unit, and 1014 is a weight calculation unit.
[0004]
The operation of the conventional imaging apparatus configured as described above will be described. From the video signal of the image sensor 1003, the first and second region luminance comparison unit 1009 obtains the vertical luminance difference p. The peak value error calculation unit 1011 calculates the peak value error between the peak value from the peak value detection unit 1010 and the target value. An average value error calculation unit 1012 calculates an average value error between the full screen average from the full screen average detection unit 1007 and the target value. A weight calculation unit 1014 calculates weight coefficients for the peak value error and the average value error from p. The weighted average error calculation unit 1013 calculates the weighted average error using the peak value error from the peak value error calculation unit 1011, the average value error from the average value error calculation unit 1012, and the weight coefficient from the weight calculation unit 1014. calculate. The aperture control unit 1008 controls the aperture opening so as to keep the weighted average error at zero.
[0005]
The first and second region luminance comparison unit 1009 obtains the upper and lower luminance difference p to determine backlight and overorder light. When the average brightness of the first area including the upper part of the screen is higher than the average brightness of the second area including the center of the screen, the backlight is determined to be backlit. At this time, the weight calculation unit 1014 further reduces the weight coefficient of the peak value error with respect to the average value error from p. Since the average value error works in the direction in which the diaphragm 1002 opens and the peak value error works in the direction in which the diaphragm 1002 closes, the diaphragm 1002 opens when the weight coefficient of the peak value error decreases. Therefore, when photographing a backlight subject, it is possible to perform appropriate exposure control by opening the aperture so that the backlight subject does not become dark.
[0006]
[Problems to be solved by the invention]
In such a conventional imaging apparatus, it is required to determine backlight and over-order light and perform appropriate exposure control.
[0007]
However, in the above-described conventional example, it is possible to determine backlighting in the case of strong backlighting. However, sufficient backlight correction is performed so that the backlighting subject is brightened only by reducing the weighting factor of the peak value error with respect to the average value error. There is a problem that it cannot be done.
[0008]
Further, in the above-described conventional example, by increasing the target value, sufficient backlight correction can be performed even in strong backlight, but there is also a problem that the aperture is excessively opened in a normal follow-up subject.
[0009]
Further, in the above-described conventional example, when overorder light is used, the average of the whole screen becomes small, and when the overorder light is determined, the weight of the average value error becomes large.
[0010]
SUMMARY OF THE INVENTION An object of the present invention is to provide an imaging apparatus capable of determining backlighting and over-ordering light and performing appropriate exposure control even for a forward-lighting subject and a strong backlighting subject.
[0011]
[Means for Solving the Problems]
  An imaging apparatus according to a first aspect of the present invention includes an area luminance detection unit that divides an entire screen area into a plurality of areas and detects a representative luminance value of each area, and a representative luminance value of each area.Each weight was set to an approximately equal weightA weighted 1 full screen average calculating unit for calculating a weighted 1 full screen average using each weight corresponding to each of the first areas, and a representative luminance value of each of the areasEach weight is set to the weight of the screen center weightA weighted 2 full screen average calculating unit that calculates a weighted 2 full screen average using each weight corresponding to each second region;
A peak value 1 detection unit for detecting a peak luminance value in the entire screen area, and a weight average of the weighted 1 full screen average value and the weighted 2 full screen average value to calculate a full screen average value And an error between the full screen average value obtained by the full screen average value calculation unit and a predetermined target value(Average value error)Between the peak luminance value obtained by the average value error calculation unit for calculating the peak value 1 and the peak value 1 detection unit and a predetermined target value(Peak value error)Obtained from the peak value error calculation unit and the average value error calculation unitSaidObtained from the average value error and the peak value error calculation unitSaidA weighted average error calculating unit that calculates a weighted average error by weighted average of peak value errors, an aperture control unit that determines an aperture opening degree so that the weighted average error follows 0, and a first area including an upper portion of the screen. Average brightnessFromAverage brightness of the second area including the center of the screenSubtract 1 The brightness difference control value (hereinafter abbreviated as CNT) between the second area and the second area is calculated, and when the brightness difference CNT value is positive, it is determined to be backlight, and the magnitude of the brightness difference CNT value is Determine the intensity of backlightThe first and second region luminance comparison units, and the first and second region luminance comparison unitsCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 full screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the luminance difference In accordance with the value of CNT, the weight of the weighted 2 full screen average value gradually increases.The weighted 1 full screen average value and the weighted 2 full screen average valueAddition ofAn average weight calculating unit that calculates a weight of a weighted average, and a first and second area luminance comparing unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight ratio between the average value error and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, That is, the weight of the average value error gradually increases according to the value of the luminance difference CNT.The average value error and the peak value errorAddition ofAnd a peak weight calculation unit for calculating the weight of the weighted average.
[0012]
According to the imaging device of the first aspect, the first and second area luminance comparison units compare the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. Backlight determination is performed. According to the comparison result, an appropriate weighted average weight of the average value error and the peak value error is calculated, and further, an appropriate weighted average weight of the weighted 1 full screen average value and the weighted 2 full screen average value is calculated, Since the aperture can be further opened, it is possible to provide an imaging apparatus that can perform back light / over-order light determination and perform appropriate exposure control even for a light-traffic subject or a strong backlight subject. Is obtained.
[0013]
  An imaging apparatus according to a second aspect includes an area luminance detection unit that divides an entire screen area into a plurality of areas and detects a representative luminance value of each area;Each weight was set to an approximately equal weightA weighted 1 full-screen average calculating unit that calculates a weighted 1 full-screen average value using the representative luminance value and each weight corresponding to each first area, and the representative luminance value of each areaEach weight is set to the weight of the screen center weightA weighting 2 full screen average calculating unit for calculating a weighted 2 full screen average value using each weight corresponding to each second region, and a peak value 1 detecting unit for detecting a peak 1 luminance value in the full screen region; A peak value 2 detector for detecting a peak 2 luminance value in a second area including the center of the screen, a peak value calculator for calculating a peak value by weighted averaging the peak 1 luminance value and the peak 2 luminance value, A full screen average value calculating unit that calculates a full screen average value by weighted averaging the weighted 1 full screen average value and the weighted 2 full screen average value, and a full screen average value obtained by the full screen average value calculating unit Between the target value and the target value(Average value error)An error between the average value error calculation unit for calculating the peak value obtained by the peak value calculation unit and a predetermined target value(Peak value error)Obtained from the peak value error calculation unit and the average value error calculation unitSaidObtained from the average value error and the peak value error calculation unitSaidA weighted average error calculating unit that calculates a weighted average error by weighted average of peak value errors, an aperture control unit that determines an aperture opening degree so that the weighted average error follows 0, and a first area including an upper portion of the screen. Average brightnessFromAverage brightness of the second area including the center of the screenThe luminance difference Control (hereinafter abbreviated as CNT) between the first region and the second region is calculated by subtraction, and when the value of the luminance difference CNT is positive, it is determined as backlight, and the luminance difference The intensity of backlight is judged by the size of the CNT valueThe first and second region luminance comparison units, and the first and second region luminance comparison unitsCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 full screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the luminance difference In accordance with the value of CNT, the weight of the weighted 2 full screen average value gradually increases.The weighted 1 full screen average value and the weighted 2 full screen average valueAddition ofAn average weight calculating unit that calculates a weight of a weighted average, and a first and second area luminance comparing unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight ratio between the average value error and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, That is, the weight of the average value error gradually increases according to the value of the luminance difference CNT.The average value error and the peak value errorAddition ofAnd a peak weight calculation unit for calculating the weight of the weighted average.
[0014]
According to the imaging device of the second aspect, the first and second area luminance comparison units compare the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. Backlight determination is performed. According to the comparison result, an appropriate weighted average weight of the average value error and the peak value error is calculated, and further, an appropriate weighted average weight of the weighted 1 full screen average value and the weighted 2 full screen average value is calculated, The aperture can be further opened. Furthermore, by calculating the peak value by weighted average of the peak 1 luminance value and the peak 2 luminance value, the peak value can be reduced when the peak value is only at the top of the screen, and the aperture can be further opened. There is an effect that it is possible to provide an imaging apparatus capable of performing appropriate exposure control by performing determination of backlight and over-order light on a forward light subject and a strong backlight subject.
[0015]
  An imaging apparatus according to a third aspect includes an area luminance detection unit that divides an entire screen area into a plurality of areas and detects a representative luminance value of each area, and a representative luminance value of each area.Each weight is set to the weight of the screen center weightA weighting 2 full screen average calculating unit for calculating a weighted 2 full screen average value using each weight corresponding to each second region, and a peak value 1 detecting unit for detecting a peak 1 luminance value in the full screen region; A peak value 2 detector for detecting a peak 2 luminance value in a second area including the center of the screen, a peak value calculator for calculating a peak value by weighted averaging the peak 1 luminance value and the peak 2 luminance value, An error between the weighted 2 full screen average value obtained by the weighted 2 full screen average calculation unit and a predetermined target value(Average value error)An error between the average value error calculation unit for calculating the peak value obtained by the peak value calculation unit and a predetermined target value(Peak value error)Obtained from the peak value error calculation unit and the average value error calculation unitSaidObtained from the average value error and the peak value error calculation unitSaidA weighted average error calculating unit that calculates a weighted average error by weighted average of peak value errors, an aperture control unit that determines an aperture opening degree so that the weighted average error follows 0, and a first area including an upper portion of the screen. Average brightnessFromAverage brightness of the second area including the center of the screenThe luminance difference Control (hereinafter abbreviated as CNT) between the first region and the second region is calculated by subtraction, and when the value of the luminance difference CNT is positive, it is determined as backlight, and the luminance difference The intensity of backlight is judged by the size of the CNT valueThe first and second region luminance comparison units, and the first and second region luminance comparison unitsCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight ratio between the average value error and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, That is, the weight of the average value error gradually increases according to the value of the luminance difference CNT.The average value error and the peak value errorAddition ofAnd a peak weight calculation unit for calculating the weight of the weighted average.
[0016]
According to the imaging device of the third aspect, the first and second area luminance comparison unit compares the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. Backlight determination is performed. By calculating the appropriate weighted average weight of the average value error and peak value error based on this comparison result, the aperture can be opened, and the peak value is calculated by weighted average of the peak 1 luminance value and the peak 2 luminance value. As a result, when the peak value is only at the top of the screen, the peak value can be reduced and the aperture can be further opened. It is possible to provide an imaging apparatus capable of performing determination and performing appropriate exposure control.
[0017]
  An imaging apparatus according to a fourth aspect includes an area luminance detection unit that divides an entire screen area into a plurality of areas and detects a representative luminance value of each area, and a representative luminance value of each area.Each weight was set to an approximately equal weightA weighted 1 full screen average calculating unit for calculating a weighted 1 full screen average using each weight corresponding to each of the first areas, and a representative luminance value of each of the areasEach weight is set to the weight of the screen center weightUsing each weight corresponding to each second region, a weighting 2 full screen average calculating unit for calculating weighted 2 full screen average value, and weighting 1 full screen average value and weighting 2 full screen average value are weighted The average of the full screen average value calculation unit for averaging and calculating the average value of the full screen, and the error between the full screen average value obtained by the full screen average value calculation unit and a predetermined target value(Average value error)An average value error calculation unit that calculates the aperture value, a diaphragm control unit that determines the aperture of the aperture so that the average value error follows 0, and an average luminance of the first area including the upper part of the screenFromAverage brightness of the second area including the center of the screenThe luminance difference Control (hereinafter abbreviated as CNT) between the first region and the second region is calculated by subtraction, and when the value of the luminance difference CNT is positive, it is determined as backlight, and the luminance difference The intensity of backlight is judged by the size of the CNT valueThe first and second region luminance comparison units, and the first and second region luminance comparison unitsCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 full screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the luminance difference In accordance with the value of CNT, the weight of the weighted 2 full screen average value gradually increases.The weighted 1 full screen average value and the weighted 2 full screen average valueAddition ofAnd an average weight calculation unit for calculating the weight of the weighted average.
[0018]
According to the imaging device of the fourth aspect, the first and second area luminance comparison units compare the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. Backlight determination is performed. By calculating an appropriate weighted average weight of the weighted 1 full-screen average value and the weighted 2 full-screen average value based on the comparison result, the aperture can be further opened. In addition, there can be obtained an effect that it is possible to provide an imaging apparatus capable of determining backlight and excessive forward light and performing appropriate exposure control.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to FIGS.
[0020]
(Embodiment 1)
A first embodiment of the present invention will be described with reference to FIGS.
[0021]
FIG. 1 shows a block diagram of an imaging apparatus according to Embodiment 1 of the present invention. In FIG. 1, an optical image of a subject formed by an optical system 101 is incident on an image sensor (CCD) 103 through a diaphragm 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 102 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0022]
Further, the output terminal of the ASP / A / D converter 104 is connected to the input terminals of the area luminance detection unit 107 and the peak value 1 detection unit 110. The area luminance detection unit 107 divides the entire screen area into a plurality of areas, and detects the representative luminance value of each area. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area as a peak value. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111. The first and second area luminance comparison unit 111 compares the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. The output ends of the first and second region luminance comparison units 111 are connected to the input ends of the average weight calculation unit 112 and the peak weight calculation unit 116. The average weight calculator 112 calculates an average weight. The peak weight calculation unit 116 calculates a peak weight. The weighted 1 full screen average calculating unit 108 and the weighted 2 full screen average calculating unit 109 calculate the weighted 1 full screen average value and the weighted 2 full screen average value, respectively. The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113. The full screen average value calculation unit 113 calculates the full screen average value. The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates an average value error. The output terminal of the peak value 1 detection unit 110 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error. The output terminals of the peak value error calculator 114, the average value error calculator 115, and the peak weight calculator 116 are connected to the input terminal of the weighted average error calculator 117. The weighted average error calculation unit 117 calculates a weighted average error. The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0023]
FIG. 2 is a schematic diagram showing area division of the entire screen area of the area luminance detection unit in the imaging apparatus of the present embodiment. In FIG. 2, 201 indicates a first area, and 202 indicates a second area.
[0024]
FIG. 3 is a configuration diagram showing weighting factors of the weighted 1 full-screen average calculating unit and the weighting 2 full-screen average calculating unit in the imaging apparatus according to the present embodiment. 3, (a) is a weighting factor of the weighted 1 full screen average value, (b) is a weighting factor of the weighting 2 full screen average value, (c) is an example 1 of the weighting factor of the weighted full screen average value, (d ) Shows an example 2 of the weighting coefficient of the weighted full-screen average value.
[0025]
FIG. 4 is a detection area diagram showing a peak detection area of the peak value 1 detection unit in the imaging apparatus of the present embodiment. In FIG. 4, 201 indicates a first area, 202 indicates a second area, and 401 indicates a full screen area including the first area and the second area.
[0026]
FIG. 5 is a relationship diagram showing the relationship between the luminance difference and weight of the first region and the second region when the full screen average value calculation unit calculates the full screen average value in the imaging apparatus of the present embodiment.
[0027]
FIG. 6 is a relationship diagram illustrating the relationship between the luminance difference and the weights of the first region and the second region when the weighted average error calculation unit calculates the weighted average error in the imaging apparatus according to the present embodiment.
[0028]
The operation of the imaging apparatus of the present embodiment configured as described above will be described below. In FIG. 1, an optical image of a subject formed by an optical system 101 is incident on an image sensor (CCD) 103 through a diaphragm 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 103 is double-sampled and gain-adjusted and then analog / digital converted. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0029]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminals of the area luminance detection unit 107 and the peak value 1 detection unit 110.
[0030]
The operation of the area luminance detection unit 107 will be described with reference to FIG. The area luminance detection unit 107 divides the entire screen area into 8 areas in the horizontal direction and 6 areas in the vertical direction as shown in FIG. In each region, the luminance signal is simply added for each pixel, and divided by an arbitrary numerical value (for example, 256) to detect the representative luminance value of each region. If an arbitrary numerical value is the number of pixels in each area, the representative luminance value becomes the average luminance value of each area. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111.
[0031]
  The operation of the first and second region luminance comparison unit 111 will be described with reference to FIG. In the first and second area luminance comparison unit 111, each area in the second area including the center of the screen is obtained from the average luminance obtained by averaging the representative luminance values in the first area including the upper part of the screen shown in FIG. Subtract the average brightness that averaged the representative brightness values of the first and second areas, and the brightness difference between the first and second areasControl (hereinafter abbreviated as CNT) valueIs calculated. Luminance difference CNTThe value of theIf becomes negative, perform negative clipping. Also, luminance difference CNTThe value of theIf the value exceeds the maximum value, the maximum value is clipped. The output ends of the first and second region luminance comparison units 111 are connected to the input ends of the average weight calculation unit 112 and the peak weight calculation unit 116.
[0032]
The operation of the weighted 1 full screen average calculator 108 and the weighted 2 full screen average calculator 109 will be described with reference to FIG. The weighted 1 full screen average calculating unit 108 multiplies the representative luminance of each area by the weighting factor of the weighted 1 full screen average value shown in FIG. 3A, and adds all the multiplication results of each area. Next, all the weighting factors of each region are added. Finally, the weighted 1 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0033]
Similarly, the weighting coefficient of the weighting 2 full screen average value shown in FIG. 3B is multiplied by the representative luminance of each area, and all the multiplication results of each area are added. Next, all the weighting factors of each region are added. Finally, the weighted 2 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0034]
The weighting factor of the weighted 2 full screen average value shown in FIG. 3B is set so that the weight becomes smaller as it goes to the upper part of the screen. When there is a bright subject with a large luminance signal at the top of the screen, weighting 2 When the average value of the entire screen is calculated by the above method, the weight at the top of the screen is small, so that the weight of the entire screen is uniform. 1 smaller than the average value of all screens. Conversely, when there is a dark subject with a small luminance signal at the top of the screen, the weighted 2 full screen average value is larger than the weighted 1 full screen average value.
[0035]
The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113.
[0036]
The operation of the peak value 1 detection unit 110 will be described with reference to FIG. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area shown in FIG. 4 as a peak value.
[0037]
The operation of the average weight calculation unit 112 will be described with reference to FIG. The average weight calculator 112 calculates the average weight α so that the luminance difference between the first area and the second area shown in FIG. The average weight α indicates the weighting factor of the weighting 2 full screen average, and the weighting factor of the weighting 1 full screen average is obtained by 100% −α. That is, when the luminance difference CNT between the first region and the second region is negative, the average weight α is set to 0% so that the weighted 1 full screen average becomes 100%. As the luminance difference CNT between the first area and the second area gradually increases, the average weight α is calculated so that the weight of the weighted 2 full screen average increases. When the luminance difference CNT between the first area and the second area reaches the maximum value, the average weight α is set to 100% so that the weighting 2 full screen average becomes 100%.
[0038]
The operation of the full screen average value calculation unit 113 will be described with reference to FIG. The full screen average value calculation unit 113 performs weighted averaging of the weighted 1 full screen average value and the weighted 2 full screen average value using the average weight α to calculate the full screen average value. Assuming that the average value of the entire screen is AVEY, the weighted 1 average value of the entire screen is AVEY1, and the weighted 2 average value of the entire screen is AVEY2, the following equation is obtained: AVEY = (AVEY1 × (100−α) + AVEY2 × α) / 100 The screen average value can be calculated.
[0039]
The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value.
[0040]
The output terminal of the peak value 1 detection unit 110 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error by subtracting the peak value from a predetermined target value.
[0041]
Here, when the opening / closing operation of the diaphragm 102 is stable, the peak value error is usually a negative value, and the average value error is a positive value. The peak value error and the average value error are weighted and canceled to cancel each other, so that the diaphragm 102 is balanced at the diaphragm position where the weighted average error is zero. Output terminals of the peak value error calculation unit 114 and the average value error calculation unit 115 are connected to an input terminal of the weighted average error calculation unit 117.
[0042]
The operation of the peak weight calculation unit 116 will be described with reference to FIG. The peak weight calculation unit 116 calculates the peak weight β so that the luminance difference between the first region and the second region shown in FIG. The peak weight β indicates the weight coefficient of the peak value error, and the weight coefficient of the average value error is obtained by 100% −β. That is, when the luminance difference CNT between the first region and the second region is negative, the peak weight β is set to 25% so that the peak value error is 25%. The peak weight β is calculated so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. When the luminance difference CNT between the first region and the second region reaches the maximum value, the peak weight β is set to 0% so that the average value error becomes 100%. The output terminal of the peak weight calculation unit 116 is connected to the input terminal of the weighted average error calculation unit 117.
[0043]
The operation of the weighted average error calculation unit 117 will be described with reference to FIG. The weighted average error calculation unit 117 uses the peak value error obtained by the peak weight calculation unit 116 as the peak value error obtained by the peak value error calculation unit 114 and the average value error obtained by the average value error calculation unit 115. To calculate a weighted average error. When the weighted average error is GOSA, the peak value error is PGOSA, and the average value error is AGOSA, the weighted average error can be calculated by the following formula: GOSA = (AGOSA × (100−β) + PGOSA × β) / 100. .
[0044]
The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0045]
The operation of the backlight determination will be described with reference to FIGS. The backlight determination is performed by the first / second region luminance comparison unit 111. The first and second area luminance comparison unit 111 subtracts the average luminance of the second area including the center of the screen from the average luminance of the first area including the upper part of the screen to obtain the luminance difference CNT between the first area and the second area. Calculate. If a main subject such as a person is photographed at the center of the screen during backlighting, the main subject will be dark and the upper part of the screen will often be a bright subject such as the sky or clouds. Therefore, the luminance difference CNT between the first region and the second region is Positive value. As the intensity of backlight increases, the value of the luminance difference CNT between the first area and the second area increases. That is, the backlight determination is performed depending on whether the luminance difference CNT between the first region and the second region is positive or negative, and the backlight intensity is determined based on the value of the luminance difference CNT. In addition, if the main subject is photographed with the main subject at the center of the screen during excessive light, the luminance difference CNT between the first region and the second region becomes a negative value because the main subject is bright.
[0046]
Next, the operation for correcting the backlight when the backlight is determined will be described with reference to FIGS. At the time of backlight determination, the luminance difference CNT between the first area and the second area calculated by the first and second area luminance comparison unit 111 is a positive value, so there are more bright subjects in the first area than in the second area. It will be. At this time, if the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is smaller than the weighted 1 full screen average value.
[0047]
Also, if the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 is a large value, the average weight calculation unit 112 determines the luminance between the first area and the second area. As the difference CNT gradually increases, the average weight α is calculated so that the weight of the weight 2 full-screen average increases. Therefore, the full screen average value calculation unit 113 calculates a full screen average value close to the weighted 2 full screen average by the weighted average. Therefore, the full screen average value is smaller than the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the full screen average value is smaller than the conventional full screen average value, the average value error becomes large.
[0048]
Further, the peak weight calculation unit 116 calculates the peak weight β so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. Therefore, the peak weight β is small during backlighting. Therefore, the weighted average error calculated by the weighted average error calculation unit 117 is a value close to the average value error. Since the average value error is large, the weighted average error is also large. When the weighted average error is large, the diaphragm control unit 118 operates in the direction of opening the diaphragm 102 so as to reduce the weighted average error. Therefore, the backlight 102 can be opened by the conventional imaging apparatus and backlight compensation can be performed during backlighting.
[0049]
Next, the operation of the aperture control when the forward light / over-forward light is determined will be described with reference to FIGS. At the time of forward light / over-order light determination, the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 becomes a negative value, so that the second area including the center of the screen is displayed. In comparison, there are many dark subjects in the first area including the upper part of the screen. At this time, if the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is larger than the weighted 1 full screen average value. Further, since the weighting factors of the weighted 1 full screen average value are equal, the weighted 1 full screen average value is equal to the full screen average value of the conventional imaging apparatus.
[0050]
If the luminance difference CNT between the first region and the second region calculated by the first / second region luminance comparison unit 111 is a negative value, negative clipping is performed. When the luminance difference CNT between the first area and the second area is 0, the average weight calculation unit 112 calculates the average weight so that the weight of the weight 1 total screen average is 100%. Therefore, the full screen average value calculation unit 113 calculates the weighted 1 full screen average as the full screen average value by the weighted average. Therefore, the full screen average value is the same as the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the average value of the entire screen is the same as the average value of the entire screen of the conventional imaging apparatus, the average value error is also equal.
[0051]
Further, the peak weight calculation unit 116 calculates the peak weight so that the peak weighting coefficient becomes the maximum value because the luminance difference CNT between the first region and the second region becomes 0 which is negatively clipped. Therefore, the peak weight is 25% during the forward light and the over light. Therefore, the weighted average error calculation unit 117 calculates the weighted average error by performing a weighted average of the average value error and the peak value error at 3: 1.
[0052]
If the maximum value of the peak weight is set equal to the setting of the conventional imaging device, the peak value error and the average value error are equal to those of the conventional imaging device, so that the weighted average error is also equal to that of the conventional imaging device. Therefore, at the time of forward light / over-forward light, control equivalent to the diaphragm control of the conventional imaging apparatus can be performed.
[0053]
Therefore, the aperture can be opened more than the conventional imaging device during backlighting, and the aperture control is the same as that of the conventional imaging device during follow-up / over-forward illumination, so an imaging device capable of performing appropriate exposure control is provided. Obtainable.
[0054]
With this configuration, the first and second area luminance comparison unit 111 performs backlight determination by comparing the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. . At the time of backlight determination, the weighted average peak weight is calculated so that the average value error weighting factor is greater than the peak value error weighting factor, and the weighting 1 full screen average weighting factor is the weighting 2 full screen average value. By calculating the average weight of the weighted average so as to be smaller than the weight coefficient, the aperture can be further opened. Furthermore, when determining the following light, it is possible to perform aperture control equivalent to that of a conventional imaging apparatus, and therefore, it is possible to determine the backlighting and the following light for the following light subject and the strong backlighting subject, and perform appropriate exposure control. The effect that the imaging device which can be provided can be provided can be obtained.
[0055]
In the above description of the operation, the entire screen area is divided into 8 horizontal areas and 6 vertical areas. The backlight determination is performed in the first area including the upper part of the screen and the second area including the center of the screen. As long as the number of divisions is possible, the same effect can be obtained even if the number of divisions on the entire screen is changed or even division is not performed.
[0056]
In the above description of the operation, the first region and the second region have been described with the region setting shown in FIG. 2. However, if the first region including the upper part of the screen and the second region including the center of the screen are used, The same effect can be obtained even if the area is set.
[0057]
In the above description of the operation, the operation is described using the luminance signal. However, the same effect can be obtained by using a signal close to the luminance signal such as a maximum signal of the RGB signal or a G signal.
[0058]
In the above description of the operation, the weighting factor of weighting 2 for all screens is the same as the weighting factor shown in FIG. 3. If the weighting factor at the top of the screen is small, the weighting factor at the center of the screen is increased. The same effect can be obtained with a weighting factor of center weight (for example, the weighting factor of the weighted full-screen average shown in FIGS. 3C and 3D).
[0059]
(Embodiment 2)
A second embodiment of the present invention will be described with reference to FIGS.
[0060]
FIG. 7 is a block diagram of the imaging apparatus according to this embodiment. 7 is different from FIG. 1 showing the block diagram of the imaging apparatus according to Embodiment 1 in that a peak value 2 detection unit 701 and a peak value calculation unit 702 are added. That is, the optical image of the subject formed by the optical system 101 enters the image sensor (CCD) 103 through the stop 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 102 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0061]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminals of the area luminance detection unit 107, the peak value 1 detection unit 110, and the peak value 2 detection unit 701. The area luminance detection unit 107 divides the entire screen area into a plurality of areas, and detects the representative luminance value of each area. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area as the peak value 1. The peak value 2 detection unit 701 detects the value of the maximum luminance signal in the second area including the center of the screen as the peak value 2. The output ends of the peak value 1 detection unit 110 and the peak value 2 detection unit 701 are connected to the input end of the peak value calculation unit 702. The peak value calculation unit 702 calculates a peak value by performing a weighted average of the peak value 1 and the peak value 2. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111. The first and second area luminance comparison unit 111 compares the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. The output ends of the first and second region luminance comparison units 111 are connected to the input ends of the average weight calculation unit 112 and the peak weight calculation unit 116. The average weight calculator 112 calculates an average weight. The peak weight calculation unit 116 calculates a peak weight. The weighted 1 full screen average calculating unit 108 and the weighted 2 full screen average calculating unit 109 calculate the weighted 1 full screen average value and the weighted 2 full screen average value, respectively. The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113. The full screen average value calculation unit 113 calculates the full screen average value. The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates an average value error. The output terminal of the peak value calculation unit 702 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error. The output terminals of the peak value error calculator 114, the average value error calculator 115, and the peak weight calculator 116 are connected to the input terminal of the weighted average error calculator 117. The weighted average error calculation unit 117 calculates a weighted average error. The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0062]
FIG. 2 is a schematic diagram showing area division of the entire screen area of the area luminance detection unit in the imaging apparatus of the present embodiment. In FIG. 2, 201 indicates a first area, and 202 indicates a second area.
[0063]
FIG. 3 is a configuration diagram showing weighting factors of the weighted 1 full-screen average calculating unit and the weighting 2 full-screen average calculating unit in the imaging apparatus according to the present embodiment. 3, (a) is a weighting factor of the weighted 1 full screen average value, (b) is a weighting factor of the weighting 2 full screen average value (c) is an example 1 of the weighting factor of the weighted full screen average value, and (d). Indicates Example 2 of the weighting coefficient of the weighted full-screen average value.
[0064]
FIG. 4 is a detection area diagram showing the peak detection areas of the peak value 1 detection unit and the peak value 2 detection unit in the imaging apparatus of the present embodiment. In FIG. 4, 201 indicates a first area, 202 indicates a second area, and 401 indicates a full screen area including the first area and the second area.
[0065]
FIG. 5 is a relationship diagram showing the relationship between the luminance difference and weight of the first region and the second region when the full screen average value calculation unit calculates the full screen average value in the imaging apparatus of the present embodiment.
[0066]
FIG. 6 is a relationship diagram illustrating the relationship between the luminance difference and the weights of the first region and the second region when the weighted average error calculation unit calculates the weighted average error in the imaging apparatus according to the present embodiment.
[0067]
The operation of the imaging apparatus of the present embodiment configured as described above will be described below. In FIG. 7, an optical image of a subject formed by the optical system 101 is incident on an image sensor (CCD) 103 through a diaphragm 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 103 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0068]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminals of the area luminance detection unit 107, the peak value 1 detection unit 110, and the peak value 2 detection unit 701.
[0069]
The operation of the area luminance detection unit 107 will be described with reference to FIG. The area luminance detection unit 107 divides the entire screen area into 8 areas in the horizontal direction and 6 areas in the vertical direction as shown in FIG. In each region, the luminance signal is simply added for each pixel, and divided by an arbitrary numerical value (for example, 256) to detect the representative luminance value of each region. If an arbitrary numerical value is the number of pixels in each area, the representative luminance value becomes the average luminance value of each area. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111.
[0070]
The operation of the first and second region luminance comparison unit 111 will be described with reference to FIG. In the first and second area luminance comparison unit 111, each area in the second area including the center of the screen is obtained from the average luminance obtained by averaging the representative luminance values in the first area including the upper part of the screen shown in FIG. The average luminance obtained by averaging the representative luminance values is subtracted to calculate the luminance difference CNT between the first region and the second region. When the luminance difference CNT becomes negative, negative clipping is performed. When the luminance difference CNT exceeds the maximum value, the maximum value clip is performed. The output ends of the first and second region luminance comparison units 111 are connected to the input ends of the average weight calculation unit 112 and the peak weight calculation unit 116.
[0071]
The operation of the weighted 1 full screen average calculator 108 and the weighted 2 full screen average calculator 109 will be described with reference to FIG. The weighted 1 full screen average calculating unit 108 multiplies the representative luminance of each area by the weighting factor of the weighted 1 full screen average value shown in FIG. 3A, and adds all the multiplication results of each area. Next, all the weighting factors of each region are added. Finally, the weighted 1 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0072]
Similarly, the weighting coefficient of the weighting 2 full screen average value shown in FIG. 3B is multiplied by the representative luminance of each area, and all the multiplication results of each area are added. Next, all the weighting factors of each region are added. Finally, the weighted 2 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0073]
The weighting factor of the weighted 2 full screen average value shown in FIG. 3B is set so that the weight becomes smaller as it goes to the upper part of the screen. When there is a bright subject with a large luminance signal at the top of the screen, weighting 2 When the average value of the entire screen is calculated by the above method, the weight at the top of the screen is small, so that the weight of the entire screen is uniform. 1 smaller than the average value of all screens. Conversely, when there is a dark subject with a small luminance signal at the top of the screen, the weighted 2 full screen average value is larger than the weighted 1 full screen average value.
[0074]
The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113.
[0075]
The operation of the peak value 1 detection unit 110 and the peak value 2 detection unit 701 will be described with reference to FIG. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area shown in FIG. The peak value 2 detection unit 701 detects the value of the maximum luminance signal in the second area including the center of the screen shown in FIG. Next, the operation of the peak value calculation unit 702 will be described. The peak value calculation unit 702 calculates a peak value by performing a weighted average of the peak value 1 detected by the peak value 1 detection unit 110 and the peak value 2 detected by the peak value 2 detection unit 701. For example, when there is a peak value at the top of the screen, the peak value is smaller than the peak value 1 when the peak value is calculated by performing a weighted average of the peak value 1 and the peak value 2 at 1: 1. If the peak value is in the center of the screen, the peak value is equal to the peak value 1 when the peak value 1 and the peak value 2 are weighted and averaged 1: 1 and the peak value is calculated.
[0076]
The operation of the average weight calculation unit 112 will be described with reference to FIG. The average weight calculator 112 calculates the average weight α so that the luminance difference between the first area and the second area shown in FIG. The average weight α indicates the weighting factor of the weighting 2 full screen average, and the weighting factor of the weighting 1 full screen average is obtained by 100% −α. That is, when the luminance difference CNT between the first region and the second region is negative, the average weight α is set to 0% so that the weighted 1 full screen average becomes 100%. As the luminance difference CNT between the first area and the second area gradually increases, the average weight α is calculated so that the weight of the weighted 2 full screen average increases. When the luminance difference CNT between the first area and the second area reaches the maximum value, the average weight α is set to 100% so that the weighting 2 full screen average becomes 100%.
[0077]
The operation of the full screen average value calculation unit 113 will be described with reference to FIG. The full screen average value calculation unit 113 performs weighted averaging of the weighted 1 full screen average value and the weighted 2 full screen average value using the average weight α to calculate the full screen average value. Assuming that the average value of the entire screen is AVEY, the weighted 1 average value of the entire screen is AVEY1, and the weighted 2 average value of the entire screen is AVEY2, the following equation is obtained: AVEY = (AVEY1 × (100−α) + AVEY2 × α) / 100 The average value of the entire screen can be calculated.
[0078]
The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value.
[0079]
The output terminal of the peak value calculation unit 702 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error by subtracting the peak value from a predetermined target value.
[0080]
Here, when the opening / closing operation of the diaphragm 102 is stable, the peak value error is usually a negative value, and the average value error is a positive value. The peak value error and the average value error are weighted and canceled to cancel each other, so that the diaphragm 102 is balanced at the diaphragm position where the weighted average error is zero. Output terminals of the peak value error calculation unit 114 and the average value error calculation unit 115 are connected to an input terminal of the weighted average error calculation unit 117.
[0081]
The operation of the peak weight calculation unit 116 will be described with reference to FIG. The peak weight calculation unit 116 calculates the peak weight β so that the luminance difference between the first region and the second region shown in FIG. The peak weight β indicates the weight coefficient of the peak value error, and the weight coefficient of the average value error is obtained by 100% −β. That is, when the luminance difference CNT between the first region and the second region is negative, the peak weight β is set to 25% so that the peak value error is 25%. The peak weight β is calculated so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. When the luminance difference CNT between the first region and the second region reaches the maximum value, the peak weight β is set to 0% so that the average value error becomes 100%. The output terminal of the peak weight calculation unit 116 is connected to the input terminal of the weighted average error calculation unit 117.
[0082]
The operation of the weighted average error calculation unit 117 will be described with reference to FIG. The weighted average error calculation unit 117 uses the peak value error obtained by the peak weight calculation unit 116 as the peak value error obtained by the peak value error calculation unit 114 and the average value error obtained by the average value error calculation unit 115. To calculate a weighted average error. When the weighted average error is GOSA, the peak value error is PGOSA, and the average value error is AGOSA, the weighted average error can be calculated by the following formula: GOSA = (AGOSA × (100−β) + PGOSA × β) / 100. it can.
[0083]
The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0084]
The operation of the backlight determination will be described with reference to FIGS. The backlight determination is performed by the first / second region luminance comparison unit 111. The first and second area luminance comparison unit 111 subtracts the average luminance of the second area including the center of the screen from the average luminance of the first area including the upper part of the screen to obtain the luminance difference CNT between the first area and the second area. Calculate. If a main subject such as a person is photographed in the center of the screen during backlighting, the main subject will be dark and the top of the screen will often be a bright subject such as a sky or a cloud. Therefore, the luminance difference CNT between the first region and the second region is Positive value. As the intensity of backlight increases, the value of the luminance difference CNT between the first area and the second area increases. That is, the backlight determination is performed depending on whether the luminance difference CNT between the first region and the second region is positive or negative, and the backlight intensity is determined based on the value of the luminance difference CNT. In addition, if the main subject is photographed with the main subject at the center of the screen during excessive light, the luminance difference CNT between the first region and the second region becomes a negative value because the main subject is bright.
[0085]
Next, the operation for correcting the backlight when the backlight is determined will be described with reference to FIGS. At the time of backlight determination, the luminance difference CNT between the first area and the second area calculated by the first and second area luminance comparison unit 111 is a positive value, so there are more bright subjects in the first area than in the second area. It will be. At this time, if the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is smaller than the weighted 1 full screen average value.
[0086]
Also, if the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 is a large value, the average weight calculation unit 112 determines the luminance between the first area and the second area. As the difference CNT gradually increases, the average weight α is calculated so that the weight of the weight 2 full-screen average increases. Therefore, the full screen average value calculation unit 113 calculates a full screen average value close to the weighted 2 full screen average by the weighted average. Therefore, the full screen average value is smaller than the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the full screen average value is smaller than the conventional full screen average value, the average value error becomes large.
[0087]
Further, since there are more bright subjects in the first area than in the second area during backlighting, the peak value 1 of the entire screen area is larger than the peak value 2 of the second area including the center of the screen. Here, when the peak value calculation unit 702 calculates the peak value by weighted averaging the peak value 1 and the peak value 2 at 1: 1, the peak value is smaller than the peak value detected by the conventional imaging device in the entire screen area. Become. Next, the peak value error calculation unit 114 subtracts the peak value from the predetermined target value to calculate the peak value error. Since the peak value is smaller than the conventional peak value, the absolute value of the peak value error is small.
[0088]
Further, the peak weight calculation unit 116 calculates the peak weight β so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. Therefore, the peak weight β is small during backlighting. Therefore, the weighted average error calculated by the weighted average error calculation unit 117 is a value close to the average value error. Since the average value error increases and the peak value error decreases, the weighted average error also increases. When the weighted average error is large, the diaphragm control unit 118 operates in the direction of opening the diaphragm 102 so as to reduce the weighted average error. Therefore, the backlight 102 can be opened by the conventional imaging apparatus and backlight compensation can be performed during backlighting.
[0089]
Next, the operation of the aperture control when the forward light / over-forward light is determined will be described with reference to FIGS. At the time of forward light / over-order light determination, the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 becomes a negative value, so that the second area including the center of the screen is displayed. In comparison, there are many dark subjects in the first area including the upper part of the screen. At this time, when the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is larger than the weighted 1 full screen average value. In addition, since the weighting factors of the weighted 1 full screen average value are equal, the weighted 1 full screen average value is equal to the full screen average value of the conventional imaging apparatus.
[0090]
If the luminance difference CNT between the first region and the second region calculated by the first / second region luminance comparison unit 111 is a negative value, negative clipping is performed. When the luminance difference CNT between the first area and the second area is 0, the average weight calculation unit 112 calculates the average weight so that the weight of the weight 1 total screen average is 100%. Therefore, the full screen average value calculation unit 113 calculates the weighted 1 full screen average as the full screen average value by the weighted average. Therefore, the full screen average value is the same as the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the average value of the entire screen is the same as the average value of the entire screen of the conventional imaging apparatus, the average value error is also equal.
[0091]
In addition, since there are more bright subjects in the second area than in the first area at the time of forward light / over-order light, the peak value 1 of the entire screen area is equal to the peak value 2 of the second area including the center of the screen. Here, when the peak value calculation unit 702 calculates the peak value by weighted averaging the peak value 1 and the peak value 2 at 1: 1, the peak value is equal to the peak value detected by the conventional imaging device in the entire screen area. Become. Next, the peak value error calculation unit 114 subtracts the peak value from the predetermined target value to calculate the peak value error. Since the peak value is equal to the conventional peak value, the peak value error is also equal.
[0092]
Further, the peak weight calculation unit 116 calculates the peak weight so that the peak weighting coefficient becomes the maximum value because the luminance difference CNT between the first region and the second region becomes 0 which is negatively clipped. Therefore, the peak weight is 25% during the forward light and the over light. Therefore, the weighted average error calculation unit 117 calculates the weighted average error by performing a weighted average of the average value error and the peak value error at 3: 1.
[0093]
If the maximum value of the peak weight is set to be equal to the setting of the conventional imaging device, the average value error and the peak value error are equal to those of the conventional imaging device, so that the weighted average error is also equal to that of the conventional imaging device. Therefore, at the time of forward light / over-forward light, control equivalent to the diaphragm control of the conventional imaging apparatus can be performed.
[0094]
Therefore, the aperture can be opened more than the conventional imaging device during backlighting, and the aperture control is the same as that of the conventional imaging device during follow-up / over-forward illumination, so an imaging device capable of performing appropriate exposure control is provided. Obtainable.
[0095]
With this configuration, the first and second area luminance comparison unit 111 performs backlight determination by comparing the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. . At the time of backlight determination, the weighted average peak weight is calculated so that the average value error weighting factor is greater than the peak value error weighting factor, and the weighting 1 full screen average weighting factor is the weighting 2 full screen average value. By calculating the average weight of the weighted average so as to be smaller than the weight coefficient, the aperture can be further opened. Further, by calculating the peak value by weighted average of the peak 1 luminance value and the peak 2 luminance value, the peak value can be reduced when the peak value is only at the upper part of the screen, and the aperture can be further opened. In addition, when determining the following light, it is possible to perform aperture control equivalent to that of a conventional imaging device, so the backlight and the following light are determined for the following light subject and the strong backlight subject, and appropriate exposure control is performed. The effect that the imaging device which can be performed can be provided is acquired.
[0096]
In the above description of the operation, the entire screen area is divided into 8 horizontal areas and 6 vertical areas. The backlight determination is performed in the first area including the upper part of the screen and the second area including the center of the screen. As long as the number of divisions is possible, the same effect can be obtained even if the number of divisions on the entire screen is changed or even division is not performed.
[0097]
In the above description of the operation, the first region and the second region have been described with the region setting shown in FIG. 2. However, if the first region including the upper part of the screen and the second region including the center of the screen are used, The same effect can be obtained even if the area is set.
[0098]
In the above description of the operation, the operation is described using the luminance signal. However, the same effect can be obtained by using a signal close to the luminance signal such as a maximum signal of the RGB signal or a G signal.
[0099]
In the above description of the operation, the weighting factor of weighting 2 for all screens is the same as the weighting factor shown in FIG. 3. If the weighting factor at the top of the screen is small, the weighting factor at the center of the screen is increased. The same effect can be obtained with a weighting factor of center weight (for example, the weighting factor of the weighted full-screen average shown in FIGS. 3C and 3D).
[0100]
(Embodiment 3)
A third embodiment of the present invention will be described with reference to FIGS. 2 to 4, 6 and 8. FIG.
[0101]
FIG. 8 is a block diagram of the imaging apparatus according to the present embodiment. 8 is different from FIG. 7 showing the block diagram of the imaging apparatus according to the second embodiment in that the weighting 1 full screen average calculation unit 108, the average weight calculation unit 112, and the full screen average value calculation unit 113 are deleted. That is. That is, the optical image of the subject formed by the optical system 101 enters the image sensor (CCD) 103 through the stop 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 102 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0102]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminals of the area luminance detection unit 107, the peak value 1 detection unit 110, and the peak value 2 detection unit 701. The area luminance detection unit 107 divides the entire screen area into a plurality of areas, and detects the representative luminance value of each area. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area as the peak value 1. The peak value 2 detection unit 701 detects the value of the maximum luminance signal in the second area including the center of the screen as the peak value 2. The output ends of the peak value 1 detection unit 110 and the peak value 2 detection unit 701 are connected to the input end of the peak value calculation unit 702. The peak value calculation unit 702 calculates a peak value by performing a weighted average of the peak value 1 and the peak value 2. The output terminal of the region luminance detection unit 107 is connected to the input terminals of the weighted two-full screen average calculation unit 109 and the first and second region luminance comparison units 111. The first and second area luminance comparison unit 111 compares the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. The output terminal of the first / second region luminance comparison unit 111 is connected to the input terminal of the peak weight calculation unit 116. The peak weight calculation unit 116 calculates a peak weight. The weighting 2 full screen average calculation unit 109 calculates a weighting 2 full screen average value. The output terminal of the weighting 2 full screen average calculation unit 109 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates an average value error. The output terminal of the peak value calculation unit 702 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error. The output terminals of the peak value error calculator 114, the average value error calculator 115, and the peak weight calculator 116 are connected to the input terminal of the weighted average error calculator 117. The weighted average error calculation unit 117 calculates a weighted average error. The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0103]
FIG. 2 is a schematic diagram showing area division of the entire screen area of the area luminance detection unit in the imaging apparatus of the present embodiment. In FIG. 2, 201 indicates a first area, and 202 indicates a second area.
[0104]
FIG. 3 is a configuration diagram showing weighting factors of the weighted 1 full-screen average calculating unit and the weighting 2 full-screen average calculating unit in the imaging apparatus according to the present embodiment. 3, (a) is a weighting factor of the weighted 1 full screen average value, (b) is a weighting factor of the weighting 2 full screen average value (c) is an example 1 of the weighting factor of the weighted full screen average value, and (d). Indicates Example 2 of the weighting coefficient of the weighted full-screen average value.
[0105]
FIG. 4 is a detection area diagram showing the peak detection areas of the peak value 1 detection unit and the peak value 2 detection unit in the imaging apparatus of the present embodiment. In FIG. 4, 201 indicates a first area, 202 indicates a second area, and 401 indicates a full screen area including the first area and the second area.
[0106]
FIG. 6 is a relationship diagram illustrating the relationship between the luminance difference and the weights of the first region and the second region when the weighted average error calculation unit calculates the weighted average error in the imaging apparatus according to the present embodiment.
[0107]
The operation of the imaging apparatus of the present embodiment configured as described above will be described below. In FIG. 8, an optical image of a subject formed by the optical system 101 is incident on an image sensor (CCD) 103 through a diaphragm 102 and is subjected to photoelectric conversion. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 103 is double-sampled and gain-adjusted and then analog / digital converted. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0108]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminals of the area luminance detection unit 107, the peak value 1 detection unit 110, and the peak value 2 detection unit 701.
[0109]
The operation of the area luminance detection unit 107 will be described with reference to FIG. The area luminance detection unit 107 divides the entire screen area into 8 areas in the horizontal direction and 6 areas in the vertical direction as shown in FIG. In each region, the luminance signal is simply added for each pixel, and divided by an arbitrary numerical value (for example, 256) to detect the representative luminance value of each region. If an arbitrary numerical value is the number of pixels in each area, the representative luminance value becomes the average luminance value of each area. The output terminal of the region luminance detection unit 107 is connected to the input terminals of the weighted two-full screen average calculation unit 109 and the first and second region luminance comparison units 111.
[0110]
The operation of the first and second region luminance comparison unit 111 will be described with reference to FIG. In the first and second area luminance comparison unit 111, each area in the second area including the center of the screen is obtained from the average luminance obtained by averaging the representative luminance values in the first area including the upper part of the screen shown in FIG. The average luminance obtained by averaging the representative luminance values is subtracted to calculate the luminance difference CNT between the first region and the second region. When the luminance difference CNT becomes negative, negative clipping is performed. When the luminance difference CNT exceeds the maximum value, the maximum value clip is performed. The output terminal of the first / second region luminance comparison unit 111 is connected to the input terminal of the peak weight calculation unit 116.
[0111]
The operation of the weighted 2 full-screen average calculation unit 109 will be described with reference to FIG. The weighted 2 full screen average calculation unit 109 multiplies the representative luminance of each area by the weighting factor of the weighted 2 full screen average value shown in FIG. 3B, and adds all the multiplication results of each area. Next, all the weighting factors of each region are added. Finally, the weighted 2 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient. The weighted 2 full screen average value is output as the full screen average value.
[0112]
The weighting factor of the weighted 2 full screen average value shown in FIG. 3B is set so that the weight becomes smaller as it goes to the upper part of the screen. When the weighted 2 full-screen average value is calculated using the weighting factor of the weighted 2 full-screen average value shown in FIG. 3B, the weight at the top of the screen becomes small when the subject has a bright luminance signal at the top of the screen and a large luminance signal. Therefore, the weight of the entire screen is uniform, which is smaller than the weighting 1 full screen average value. On the other hand, when there is a dark subject with a small luminance signal at the top of the screen, the weighted 1 is larger than the average value of the entire screen.
[0113]
The operation of the peak value 1 detection unit 110 and the peak value 2 detection unit 701 will be described with reference to FIG. The peak value 1 detection unit 110 detects the value of the maximum luminance signal in the entire screen area shown in FIG. The peak value 2 detection unit 701 detects the value of the maximum luminance signal in the second area including the center of the screen shown in FIG. Next, the operation of the peak value calculation unit 702 will be described. The peak value calculation unit 702 calculates a peak value by performing a weighted average of the peak value 1 detected by the peak value 1 detection unit 110 and the peak value 2 detected by the peak value 2 detection unit 701. For example, when there is a peak value at the top of the screen, the peak value is smaller than the peak value 1 when the peak value is calculated by performing a weighted average of the peak value 1 and the peak value 2 at 1: 1. If the peak value is in the center of the screen, the peak value is equal to the peak value 1 when the peak value 1 and the peak value 2 are weighted and averaged 1: 1 and the peak value is calculated.
[0114]
The output terminal of the weighting 2 full screen average calculation unit 109 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value.
[0115]
The output terminal of the peak value calculation unit 702 is connected to the input terminal of the peak value error calculation unit 114. The peak value error calculation unit 114 calculates a peak value error by subtracting the peak value from a predetermined target value.
[0116]
Here, when the opening / closing operation of the diaphragm 102 is stable, the peak value error is usually a negative value, and the average value error is a positive value. The peak value error and the average value error are weighted and canceled to cancel each other, so that the diaphragm 102 is balanced at the diaphragm position where the weighted average error is zero. Output terminals of the peak value error calculation unit 114 and the average value error calculation unit 115 are connected to an input terminal of the weighted average error calculation unit 117.
[0117]
The operation of the peak weight calculation unit 116 will be described with reference to FIG. The peak weight calculation unit 116 calculates the peak weight β so that the luminance difference between the first region and the second region shown in FIG. The peak weight β indicates the weight coefficient of the peak value error, and the weight coefficient of the average value error is obtained by 100% −β. That is, when the luminance difference CNT between the first region and the second region is negative, the peak weight β is set to 25% so that the peak value error is 25%. The peak weight β is calculated so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. When the luminance difference CNT between the first region and the second region reaches the maximum value, the peak weight β is set to 0% so that the average value error becomes 100%. The output terminal of the peak weight calculation unit 116 is connected to the input terminal of the weighted average error calculation unit 117.
[0118]
The operation of the weighted average error calculation unit 117 will be described with reference to FIG. The weighted average error calculation unit 117 uses the peak value error obtained by the peak weight calculation unit 116 as the peak value error obtained by the peak value error calculation unit 114 and the average value error obtained by the average value error calculation unit 115. To calculate a weighted average error. When the weighted average error is GOSA, the peak value error is PGOSA, and the average value error is AGOSA, the weighted average error can be calculated by the following formula: GOSA = (AGOSA × (100−β) + PGOSA × β) / 100. it can.
[0119]
The output end of the weighted average error calculation unit 117 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0120]
The backlight determination operation will be described with reference to FIGS. 2 to 4 and FIG. The backlight determination is performed by the first / second region luminance comparison unit 111. The first and second area luminance comparison unit 111 subtracts the average luminance of the second area including the center of the screen from the average luminance of the first area including the upper part of the screen to obtain the luminance difference CNT between the first area and the second area. Calculate. If a main subject such as a person is photographed at the center of the screen during backlighting, the main subject will be dark and the upper part of the screen will often be a bright subject such as the sky or clouds. Therefore, the luminance difference CNT between the first region and the second region is Positive value. As the intensity of backlight increases, the value of the luminance difference CNT between the first area and the second area increases. That is, the backlight determination is performed depending on whether the luminance difference CNT between the first region and the second region is positive or negative, and the backlight intensity is determined based on the value of the luminance difference CNT. In addition, if the main subject is photographed with the main subject at the center of the screen during excessive light, the luminance difference CNT between the first region and the second region becomes a negative value because the main subject is bright.
[0121]
Next, the operation for correcting the backlight when the backlight is determined will be described with reference to FIGS. 2 to 4 and FIGS. 6 and 8. FIG. At the time of backlight determination, the luminance difference CNT between the first area and the second area calculated by the first and second area luminance comparison unit 111 is a positive value, so there are more bright subjects in the first area than in the second area. It will be. At this time, when the weighted 2 full screen average value is calculated by using the weighted 2 full screen average value weighting factor shown in FIG. Since the weighting coefficient becomes smaller toward the top of the screen, the weighted 2 full screen average value becomes smaller than the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the full screen average value is smaller than the conventional full screen average value, the average value error becomes large.
[0122]
Further, since there are more bright subjects in the first area than in the second area during backlighting, the peak value 1 of the entire screen area is larger than the peak value 2 of the second area including the center of the screen. Here, when the peak value calculation unit 702 weights and averages the peak value 1 and the peak value 2 at 1: 1, the peak value is smaller than the peak value detected in the conventional full screen area. Next, the peak value error calculation unit 114 subtracts the peak value from the predetermined target value to calculate the peak value error. Since the peak value is smaller than the conventional peak value, the absolute value of the peak value error is small.
[0123]
Further, the peak weight calculation unit 116 calculates the peak weight β so that the peak weight coefficient decreases as the luminance difference CNT between the first region and the second region gradually increases. Therefore, the peak weight β is small during backlighting. Therefore, the weighted average error calculated by the weighted average error calculation unit 117 is a value close to the average value error. Since the average value error increases and the peak value error decreases, the weighted average error also increases. When the weighted average error is large, the diaphragm control unit 118 operates in the direction of opening the diaphragm 102 so as to reduce the weighted average error. Therefore, the backlight 102 can be opened by the conventional imaging apparatus and backlight compensation can be performed during backlighting.
[0124]
Next, the operation of the aperture control when the over-order light determination is made will be described with reference to FIGS. 2 to 4, 6, and 8. At the time of forward light / over-order light determination, the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 becomes a negative value, so that the second area including the center of the screen is displayed. In comparison, there are many dark subjects in the first area including the upper part of the screen. At this time, when the weighted 2 full screen average value is calculated by the weighted 2 full screen average calculation unit 109 using the weighting factor of the weighted 2 full screen average value shown in FIG. Since the weight coefficient is smaller as the coefficient is at the top of the screen, the weighted 2 full-screen average value is larger than the full-screen average value of the conventional imaging apparatus. Since the average value of the entire screen is larger than the average value of the entire screen of the conventional imaging device, the average value error is reduced.
[0125]
In addition, since there are more bright subjects in the second area than in the first area during over-order light, the peak value 1 of the entire screen area is equal to the peak value 2 of the second area including the center of the screen. Here, when the peak value calculation unit 702 calculates the peak value by weighted averaging the peak value 1 and the peak value 2 at 1: 1, the peak value is equal to the peak value detected by the conventional imaging device in the entire screen area. Become. Next, the peak value error calculation unit 114 subtracts the peak value from the predetermined target value to calculate the peak value error. Since the peak value is equal to the peak value of the conventional imaging device, the peak value error is also equal.
[0126]
Further, the peak weight calculation unit 116 calculates the peak weight so that the peak weighting coefficient becomes the maximum value because the luminance difference CNT between the first region and the second region becomes 0 which is negatively clipped. Therefore, the peak weight is 25% during the forward light and the over light. Therefore, the weighted average error calculation unit 117 calculates the weighted average error by performing a weighted average of the average value error and the peak value error at 3: 1.
[0127]
If the maximum value of the peak weight is set equal to the setting of the conventional imaging device, the average value error is smaller than that of the conventional imaging device, and the peak value error is equal, so that the weighted average error is smaller than that of the conventional imaging device. Therefore, it is possible to perform aperture control that closes the aperture more than in the conventional imaging device during excessive forward light.
[0128]
Therefore, the aperture can be opened more than the conventional imaging device during backlighting, and the aperture control is more closed than the conventional imaging device during over-order light. Therefore, it is possible to obtain an imaging device that can suppress excessive opening of the image and perform appropriate exposure control.
[0129]
With this configuration, the first and second area luminance comparison unit 111 performs backlight determination by comparing the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. . At the time of backlight determination, the aperture can be opened by calculating the weighted average peak weight so that the average value error weight coefficient is larger than the peak value error weight coefficient. Further, by calculating the peak value by weighted average of the peak 1 luminance value and the peak 2 luminance value, the peak value can be reduced when the peak value is only at the upper part of the screen, and the aperture can be further opened. In addition, since it is possible to prevent the aperture from being opened too much at the time of over-order light determination, it is possible to determine back-light and front-light and perform appropriate exposure control even for over-order light subjects and strong backlight subjects. The effect that the imaging device which can be provided can be provided can be obtained.
[0130]
In the above description of the operation, the entire screen area is divided into 8 horizontal areas and 6 vertical areas. The backlight determination is performed in the first area including the upper part of the screen and the second area including the center of the screen. As long as the number of divisions is possible, the same effect can be obtained even if the number of divisions on the entire screen is changed or even division is not performed.
[0131]
In the above description of the operation, the first region and the second region have been described with the region setting shown in FIG. 2. However, if the first region including the upper part of the screen and the second region including the center of the screen are used, The same effect can be obtained even if the area is set.
[0132]
In the above description of the operation, the operation is described using the luminance signal. However, the same effect can be obtained by using a signal close to the luminance signal such as a maximum signal of the RGB signal or a G signal.
[0133]
In the above description of the operation, the weighting factor of weighting 2 for all screens is the same as the weighting factor shown in FIG. 3. If the weighting factor at the top of the screen is small, the weighting factor at the center of the screen is increased. The same effect can be obtained with a weighting factor of center weight (for example, the weighting factor of the weighted full-screen average shown in FIGS. 3C and 3D).
[0134]
(Embodiment 4)
Embodiment 4 of the present invention will be described with reference to FIG. 2, FIG. 3, FIG. 5 and FIG.
[0135]
FIG. 9 is a block diagram of the imaging apparatus according to this embodiment. 9 differs from FIG. 1 showing the block diagram of the imaging apparatus according to Embodiment 1 in that a peak value 1 detection unit 110, a peak value error calculation unit 114, a peak weight calculation unit 116, and a weighted average error calculation. The part 117 has been deleted. That is, the optical image of the subject formed by the optical system 101 enters the image sensor (CCD) 103 through the stop 102 and is photoelectrically converted. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 102 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0136]
The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the area luminance detecting unit 107. The area luminance detection unit 107 divides the entire screen area into a plurality of areas, and detects the representative luminance value of each area. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111. The first and second area luminance comparison unit 111 compares the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. The output terminal of the first and second area luminance comparison unit 111 is connected to the input terminal of the average weight calculation unit 112. The average weight calculator 112 calculates an average weight. The weighted 1 full screen average calculating unit 108 and the weighted 2 full screen average calculating unit 109 calculate the weighted 1 full screen average value and the weighted 2 full screen average value, respectively. The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113. The full screen average value calculation unit 113 calculates the full screen average value. The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates an average value error. The output end of the average value error calculation unit 115 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0137]
FIG. 2 is a schematic diagram showing area division of the entire screen area of the area luminance detection unit in the imaging apparatus of the present embodiment. In FIG. 2, 201 indicates a first area, and 202 indicates a second area.
[0138]
FIG. 3 is a configuration diagram showing weighting factors of the weighted 1 full-screen average calculating unit and the weighting 2 full-screen average calculating unit in the imaging apparatus according to the present embodiment. 3, (a) is a weighting factor of the weighted 1 full screen average value, (b) is a weighting factor of the weighting 2 full screen average value (c) is an example 1 of the weighting factor of the weighted full screen average value, and (d). Indicates Example 2 of the weighting coefficient of the weighted full-screen average value.
[0139]
FIG. 5 is a relationship diagram showing the relationship between the luminance difference and weight of the first region and the second region when the full screen average value calculation unit calculates the full screen average value in the imaging apparatus of the present embodiment.
[0140]
The operation of the imaging apparatus of the present embodiment configured as described above will be described below. In FIG. 9, an optical image of a subject formed by the optical system 101 is incident on an image sensor (CCD) 103 through a diaphragm 102 and subjected to photoelectric conversion. The output terminal of the image sensor 103 is connected to the input terminal of the ASP / A / D converter 104, and the output of the image sensor 103 is subjected to analog / digital conversion after double sampling and gain adjustment. The output terminal of the ASP / A / D converter 104 is connected to the input terminal of the signal processing circuit 105. The signal processing circuit 105 performs signal processing such as edge enhancement and outputs a video signal from the output terminal 106.
[0141]
The output terminal of the ASP / A / D converter 104 is also connected to the input terminal of the area luminance detection unit 107.
[0142]
The operation of the area luminance detection unit 107 will be described with reference to FIG. The area luminance detection unit 107 divides the entire screen area into 8 areas in the horizontal direction and 6 areas in the vertical direction as shown in FIG. In each region, the luminance signal is simply added for each pixel, and divided by an arbitrary numerical value (for example, 256) to detect the representative luminance value of each region. If an arbitrary numerical value is the number of pixels in each area, the representative luminance value becomes the average luminance value of each area. The output terminal of the area luminance detecting unit 107 is connected to the input terminals of the weighted 1 full screen average calculating unit 108, the weighted 2 full screen average calculating unit 109, and the first and second area luminance comparing units 111.
[0143]
The operation of the first and second region luminance comparison unit 111 will be described with reference to FIG. In the first and second area luminance comparison unit 111, each area in the second area including the center of the screen is obtained from the average luminance obtained by averaging the representative luminance values in the first area including the upper part of the screen shown in FIG. Then, the average luminance obtained by averaging the representative luminance values is subtracted to calculate the luminance difference CNT between the first region and the second region. When the luminance difference CNT becomes negative, negative clipping is performed. When the luminance difference CNT exceeds the maximum value, the maximum value clip is performed. The output terminal of the first and second area luminance comparison unit 111 is connected to the input terminal of the average weight calculation unit 112.
[0144]
The operation of the weighted 1 full screen average calculator 108 and the weighted 2 full screen average calculator 109 will be described with reference to FIG. The weighted 1 full screen average calculating unit 108 multiplies the representative luminance of each area by the weighting factor of the weighted 1 full screen average value shown in FIG. 3A, and adds all the multiplication results of each area. Next, all the weighting factors of each region are added. Finally, the weighted 1 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0145]
Similarly, the weighting coefficient of the weighting 2 full screen average value shown in FIG. 3B is multiplied by the representative luminance of each area, and all the multiplication results of each area are added. Next, all the weighting factors of each region are added. Finally, the weighted 2 full screen average value is calculated by dividing the total addition value of the multiplication results by the total addition value of the weighting coefficient.
[0146]
The weighting factor of the weighted 2 full screen average value shown in FIG. 3B is set so that the weight becomes smaller as it goes to the upper part of the screen. When there is a bright subject with a large luminance signal at the top of the screen, the weighting 2 is calculated so that the weight of the whole screen is uniform because the weight on the top of the screen is small when the average value of the whole screen is calculated by the above method. Smaller than the average value of the entire screen. Conversely, when there is a dark subject with a small luminance signal at the top of the screen, the weighted 2 full screen average value is larger than the weighted 1 full screen average value.
[0147]
The output terminals of the weighting 1 full screen average calculation unit 108, the weighting 2 full screen average calculation unit 109, and the average weight calculation unit 112 are connected to the input terminals of the full screen average value calculation unit 113.
[0148]
The operation of the average weight calculation unit 112 will be described with reference to FIG. The average weight calculator 112 calculates the average weight α so that the luminance difference between the first area and the second area shown in FIG. The average weight α indicates the weighting factor of the weighting 2 full screen average, and the weighting factor of the weighting 1 full screen average is obtained by 100% −α. That is, when the luminance difference CNT between the first region and the second region is negative, the average weight α is set to 0% so that the weighted 1 full screen average becomes 100%. As the luminance difference CNT between the first area and the second area gradually increases, the average weight α is calculated so that the weight of the weighted 2 full screen average increases. When the luminance difference CNT between the first area and the second area reaches the maximum value, the average weight α is set to 100% so that the weighting 2 full screen average becomes 100%.
[0149]
The operation of the full screen average value calculation unit 113 will be described with reference to FIG. The full screen average value calculation unit 113 performs weighted averaging of the weighted 1 full screen average value and the weighted 2 full screen average value using the average weight α to calculate the full screen average value. Assuming that the average value of the entire screen is AVEY, the weighted 1 average value of the entire screen is AVEY1, and the weighted 2 average value of the entire screen is AVEY2, the following equation is obtained: AVEY = (AVEY1 × (100−α) + AVEY2 × α) / 100 The screen average value can be calculated.
[0150]
The output terminal of the full screen average value calculation unit 113 is connected to the input terminal of the average value error calculation unit 115. The average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value.
[0151]
The output end of the average value error calculation unit 115 is connected to the input end of the aperture control unit 118. The aperture control unit 118 determines the aperture of the aperture so that the weighted average error follows 0.
[0152]
The operation of the backlight determination will be described with reference to FIGS. The backlight determination is performed by the first / second region luminance comparison unit 111. The first and second area luminance comparison unit 111 subtracts the average luminance of the second area including the center of the screen from the average luminance of the first area including the upper part of the screen to obtain the luminance difference CNT between the first area and the second area. Calculate. If a main subject such as a person is photographed at the center of the screen during backlighting, the main subject will be dark and the upper part of the screen will often be a bright subject such as the sky or clouds. Therefore, the luminance difference CNT between the first region and the second region is Positive value. As the intensity of backlight increases, the value of the luminance difference CNT between the first area and the second area increases. That is, the backlight determination is performed depending on whether the luminance difference CNT between the first region and the second region is positive or negative, and the backlight intensity is determined based on the value of the luminance difference CNT. In addition, if the main subject is photographed with the main subject at the center of the screen during excessive light, the luminance difference CNT between the first region and the second region becomes a negative value because the main subject is bright.
[0153]
Next, the operation for correcting the backlight when the backlight is determined will be described with reference to FIG. 2, FIG. 3, FIG. 5, and FIG. At the time of backlight determination, the luminance difference CNT between the first area and the second area calculated by the first and second area luminance comparison unit 111 is a positive value, so there are more bright subjects in the first area than in the second area. It will be. At this time, if the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is smaller than the weighted 1 full screen average value.
[0154]
Also, if the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 is a large value, the average weight calculation unit 112 determines the luminance between the first area and the second area. As the difference CNT gradually increases, the average weight α is calculated so that the weight of the weight 2 full-screen average increases. Therefore, the full screen average value calculation unit 113 calculates a full screen average value close to the weighted 2 full screen average by the weighted average. Therefore, the full screen average value is smaller than the conventional full screen average value. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the full screen average value is smaller than the conventional full screen average value, the average value error becomes large.
[0155]
If the average value error is large, the diaphragm control unit 118 operates in the direction to open the diaphragm 102 so as to reduce the average value error. Therefore, when the backlight is backlit, the diaphragm 102 can be opened by a conventional imaging device to perform backlight correction.
[0156]
Next, the operation of the aperture control when the forward light / over-forward light is determined will be described with reference to FIG. 2, FIG. 3, FIG. 5, and FIG. At the time of forward light / over-order light determination, the luminance difference CNT between the first area and the second area calculated by the first / second area luminance comparison unit 111 becomes a negative value, so that the second area including the center of the screen is displayed. In comparison, there are many dark subjects in the first area including the upper part of the screen. At this time, if the weighting 1 full screen average value and the weighting 2 full screen average value are calculated, the weighting factors of the weighting 1 full screen average value are equal, and the weighting factor of the weighting 2 full screen average value is weighted toward the top of the screen. Since the coefficient is small, the weighted 2 full screen average value is larger than the weighted 1 full screen average value. Further, since the weighting factors of the weighted 1 full screen average value are equal, the weighted 1 full screen average value is equal to the full screen average value of the conventional imaging apparatus.
[0157]
If the luminance difference CNT between the first region and the second region calculated by the first / second region luminance comparison unit 111 is a negative value, negative clipping is performed. When the luminance difference CNT between the first area and the second area is 0, the average weight calculation unit 112 calculates the average weight so that the weight of the weight 1 total screen average is 100%. Therefore, the full screen average value calculation unit 113 calculates the weighted 1 full screen average as the full screen average value by the weighted average. Therefore, the full screen average value is the same as the full screen average value of the conventional imaging apparatus. Next, the average value error calculation unit 115 calculates the average value error by subtracting the entire screen average value from the predetermined target value. Since the average value of the entire screen is the same as the average value of the entire screen of the conventional imaging apparatus, the average value error is also equal. Therefore, at the time of forward light / over-forward light, control equivalent to the diaphragm control of the conventional imaging apparatus can be performed.
[0158]
Therefore, the aperture can be opened more than the conventional imaging device during backlighting, and the aperture control is the same as that of the conventional imaging device during follow-up / over-forward illumination, so an imaging device capable of performing appropriate exposure control is provided. Obtainable.
[0159]
With this configuration, the first and second area luminance comparison unit 111 performs backlight determination by comparing the average luminance of the first area including the upper part of the screen with the average luminance of the second area including the center of the screen. . At the time of backlight determination, the aperture can be further opened by calculating the weighted average weight so that the weighting factor of the weighted 1 full screen average value is smaller than the weighting factor of the weighted 2 full screen average value. In addition, when determining the following light, it is possible to perform aperture control equivalent to that of a conventional imaging device, so the backlight and the following light are determined for the following light subject and the strong backlight subject, and appropriate exposure control is performed. The effect that the imaging device which can be performed can be provided is acquired.
[0160]
In the above description of the operation, the entire screen area is divided into 8 horizontal areas and 6 vertical areas. The backlight determination is performed in the first area including the upper part of the screen and the second area including the center of the screen. As long as the number of divisions is possible, the same effect can be obtained even if the number of divisions on the entire screen is changed or even division is not performed.
[0161]
In the above description of the operation, the first region and the second region have been described with the region setting shown in FIG. 2. However, if the first region including the upper part of the screen and the second region including the center of the screen are used, The same effect can be obtained even if the area is set.
[0162]
In the above description of the operation, the operation is described using the luminance signal. However, the same effect can be obtained by using a signal close to the luminance signal such as a maximum signal of the RGB signal or a G signal.
[0163]
In the above description of the operation, the weighting factor of weighting 2 for all screens is the same as the weighting factor shown in FIG. 3. If the weighting factor at the top of the screen is small, the weighting factor at the center of the screen is increased. The same effect can be obtained with a weighting factor of center weight (for example, the weighting factor of the weighted full-screen average shown in FIGS. 3C and 3D).
[0164]
【The invention's effect】
  As described above, according to the imaging apparatus of the first invention of the present invention, the area luminance detection unit that divides the entire screen area into a plurality of areas and detects the representative luminance value of each area;SaidRepresentative luminance value of each areaEach weight was set to an approximately equal weightA weighted 1 full screen average calculating unit for calculating a weighted 1 full screen average using each weight corresponding to each first region;SaidRepresentative luminance value of each areaEach weight is set to the weight of the screen center weightA weighting 2 full-screen average calculating unit that calculates a weighted 2 full-screen average value using each weight corresponding to each second region, a peak value 1 detecting unit that detects a peak luminance value in the full-screen region,SaidWeighting 1 full screen average value andSaidA full screen average value calculation unit for calculating a full screen average value by weighted average of the weighted 2 full screen average values;SaidError between the full screen average value obtained by the full screen average value calculation unit and the predetermined target value(Average value error)An average value error calculation unit for calculatingSaidPeak obtained by the peak value 1 detectorLuminanceBetween the value and the specified target value(Peak value error)A peak value error calculation unit for calculatingSaidObtained from the mean value error calculatorSaidMean value error andSaidObtained from the peak value error calculatorSaidA weighted average error calculation unit that calculates a weighted average error by weighted average of peak value errors;SaidA diaphragm control unit for determining a diaphragm opening so that the weighted average error follows 0, and an average luminance of the first area including the upper part of the screenFromAverage brightness of the second area including the center of the screenSubtract 1 The brightness difference CNT value of the area 2 and the second area is calculated, and when the brightness difference CNT value is positive, it is determined to be backlight, and the backlight intensity is determined based on the magnitude of the brightness difference CNT value.A first and second area luminance comparison unit,SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 full screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the luminance difference In accordance with the value of CNT, the weight of the weighting 2 full screen average value gradually increases.Weighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofAn average weight calculation unit for calculating a weight of a multiple average;SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight ratio between the average value error and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, That is, the weight of the average value error gradually increases according to the value of the luminance difference CNT.Mean value error andSaidPeak value errorAddition ofA peak weight calculator for calculating the weight of the weighted average,SaidFirst and second area luminance comparison sectionInIncluding top of screenSaidAverage brightness of the first areaFromAverage brightness of the second area including the center of the screenSubtract 1 The value of the luminance difference CNT between the area and the second area is calculated, and depending on whether the value of the luminance difference CNT is positive or negativeBacklight determination is performed. thisCalculationDepending on the resultSaidMean value error andSaidPeak value errorAddition ofCalculate the weight of the weighted average, andThe aboveWeighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofThe aperture can be further opened by calculating the weight of the weighted average.RuWith this configuration, backlight and over-order subjects are also determined for front-lit subjects and strong-backlit subjects. By performing the same aperture control as that of a conventional imaging device, an effect of providing an imaging device capable of performing appropriate exposure control can be obtained.
[0165]
  Further, according to the imaging device of the second invention of the present invention, an area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;SaidRepresentative luminance value of each areaEach weight was set to an approximately equal weightA weighted 1 full screen average calculating unit for calculating a weighted 1 full screen average using each weight corresponding to each first region;SaidRepresentative luminance value of each areaEach weight is set to the weight of the screen center weightSecondeachUsing each weight corresponding to the area, a weighting 2 full screen average calculating unit for calculating a weighting 2 full screen average value, a peak value 1 detecting unit for detecting a peak 1 luminance value in the full screen area, and a screen center A peak value 2 detector for detecting a peak 2 luminance value of the second region including;SaidPeak 1 luminance value andSaidA peak value calculation unit for calculating a peak value by weighted averaging of peak 2 luminance values;SaidA full-screen average value calculating unit that calculates a full-screen average value by weighted average of the weighted 1 full-screen average value and the weighted 2 full-screen average value;SaidError between the full screen average value obtained by the full screen average value calculation unit and the predetermined target value(Average value error)An average value error calculation unit for calculatingSaidError between the peak value obtained by the peak value calculator and the specified target value(Peak value error)A peak value error calculation unit for calculatingSaidObtained from the mean value error calculatorSaidMean value error andSaidObtained from the peak value error calculatorSaidA weighted average error calculation unit that calculates a weighted average error by weighted average of peak value errors;SaidA diaphragm control unit for determining a diaphragm opening so that the weighted average error follows 0, and an average luminance of the first area including the upper part of the screenFromAverage brightness of the second area including the center of the screenThe value of the luminance difference CNT between the first region and the second region is calculated by subtraction, and when the value of the luminance difference CNT is positive, it is determined that the light is backlit, and the magnitude of the value of the luminance difference CNT is Determine the intensity of the backlightA first and second area luminance comparison unit,SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 full screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the luminance difference In accordance with the value of CNT, the weight of the weighting 2 full screen average value gradually increases.Weighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofAn average weight calculation unit for calculating a weight of a multiple average;SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight ratio between the average value error and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, That is, the weight of the average value error gradually increases according to the value of the luminance difference CNT.Mean value error andSaidPeak value errorAddition ofA peak weight calculator for calculating the weight of the weighted average,SaidFirst and second area luminance comparison sectionInIncluding top of screenSaidAverage brightness of the first areaFromAverage brightness of the second area including the center of the screenSubtract 1 The value of the luminance difference CNT between the region and the second region is calculated, and depending on whether the value of the luminance difference CNT is positive or negativeBacklight determination is performed. thisCalculationDepending on the resultSaidMean value error andSaidPeak value errorAddition ofCalculate the weight of the weighted average, andSaidWeighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofThe aperture can be further opened by calculating the weight of the weighted average. Furthermore, by calculating the peak value by weighted average of the peak 1 luminance value and the peak 2 luminance value, the peak value can be reduced when the peak value is only at the top of the screen, and the aperture can be further opened. With this configuration, backlight and over-order subjects are also determined for front-lit subjects and strong-backlit subjects, and back-light correction is performed to open the aperture from a conventional imaging device when back-light is determined. By performing the same aperture control as that of the imaging apparatus, it is possible to provide an imaging apparatus capable of performing appropriate exposure control.
[0166]
  Further, according to the imaging device of the third invention of the present invention, an area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;SaidRepresentative luminance value of each areaEach weight is set to the weight of the screen center weightA weighting 2 full-screen average calculating unit for calculating a weighting 2 full-screen average using each weight corresponding to each second region, and a peak value 1 detecting unit for detecting a peak 1 luminance value in the full-screen region; A peak value 2 detector for detecting a peak 2 luminance value in the second region including the center of the screen;SaidPeak 1 luminance value andSaidA peak value calculation unit for calculating a peak value by weighted averaging of peak 2 luminance values;
  An error between the weighted 2 full screen average value obtained by the weighted 2 full screen average calculation unit and a predetermined target value(Average value error)An average value error calculation unit for calculatingSaidError between the peak value obtained by the peak value calculator and the specified target value(Peak value error)A peak value error calculation unit for calculatingSaidObtained from the mean value error calculatorSaidObtained from the average value error and the peak value error calculation unitSaidA weighted average error calculation unit for calculating a weighted average error by weighted average of peak value errors;SaidA diaphragm control unit for determining a diaphragm opening degree so as to follow the weighted average error to 0, and an average luminance of the first area including the upper part of the screenFromAverage brightness of the second area including the center of the screenThe value of the luminance difference CNT between the first region and the second region is calculated by subtraction, and when the value of the luminance difference CNT is positive, it is determined that the light is backlit, and the magnitude of the value of the luminance difference CNT is Determine the intensity of the backlightA first and second area luminance comparison unit,SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the ratio of the weight of the average value error of the entire screen and the peak value error is 3: 1. When the value of the luminance difference CNT is positive, In accordance with the intensity, that is, the weight of the average value error gradually increases according to the value of the value of the luminance difference CNT.Mean value error andSaidPeak value errorAddition ofA peak weight calculator for calculating the weight of the weighted average,SaidFirst and second area luminance comparison sectionInIncluding top of screenSaidAverage brightness of the first areaFromAverage brightness of the second area including the center of the screenSubtract 1 The value of the luminance difference CNT between the region and the second region is calculated, and depending on whether the value of the luminance difference CNT is positive or negativeBacklight determination is performed. thisCalculationDepending on the resultSaidMean value error andSaidPeak value errorAddition ofBy calculating the weight of the weighted average, the aperture can be opened, andSaidPeak 1 luminance value andSaidBy calculating the peak value by weighted average of the peak 2 luminance values, the peak value can be reduced when the peak value is only at the top of the screen, and the aperture can be further opened. Even for light subjects and strong backlight subjects, backlight and excessive forward light are determined. Backlight determination is performed using backlight correction that opens the diaphragm more than conventional imaging devices. By performing control to close the aperture slightly, it is possible to provide an imaging apparatus capable of performing appropriate exposure control.
[0167]
  Further, according to the imaging device of the fourth invention of the present invention, an area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;SaidRepresentative luminance value of each areaEach weight was set to an approximately equal weightA weighted 1 full screen average calculating unit for calculating a weighted 1 full screen average using each weight corresponding to each first region;SaidRepresentative luminance value of each areaEach weight is set to the weight of the screen center weightA weighted 2 full screen average calculating unit that calculates a weighted 2 full screen average using each weight corresponding to each second region;SaidThe weighted 1 full screen average value and the weighted 2 full screen average value are weighted averaged to calculate the full screen average value.Full screen average value calculatorWhen,SaidError between the full screen average value obtained by the full screen average value calculation unit and the predetermined target value(Average value error)An average value error calculation unit for calculatingSaidAn aperture control unit that determines the aperture of the aperture so that the average value error follows 0, and the average brightness of the first area including the upper part of the screenFromAverage brightness of the second area including the center of the screenThe luminance difference control (hereinafter abbreviated as CNT) between the first area and the second area is calculated by subtraction, and if the value of the luminance difference CNT is positive, it is determined as backlight, and the luminance difference CNT The intensity of the backlight is judged by the magnitude of the valueA first and second area luminance comparison unit,SaidOf the first and second area luminance comparison unitCalculationFrom the results,When the value of the luminance difference CNT is negative, the weight of the weight 1 whole screen average is 100%, and when the value of the luminance difference CNT is positive, according to the backlight intensity, that is, the value of the luminance difference CNT. In order to gradually increase the weight of the weighting 2 full-screen average value according to the size ofWeighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofAn average weight calculation unit for calculating the weight of the weighted average,SaidFirst and second area luminance comparison sectionInIncluding top of screenSaidAverage brightness of the first areaFromIncluding the center of the screenSaidThe average brightness of the second regionSubtract 1 The value of the luminance difference CNT between the area and the second area is calculated, and depending on whether the value of the luminance difference CNT is positive or negativeBacklight determination is performed. thisCalculationDepending on the resultSaidWeighting 1 full screen average value andSaidWeighted 2 screen average valueAddition ofBy calculating the weight of the weighted average, it is possible to open the aperture further, so that backlight and over-backlight subjects are also determined for backlight and strong backlight subjects. An effect of providing an imaging apparatus capable of performing appropriate exposure control by performing backlight correction for opening the diaphragm from a conventional imaging apparatus and performing the same aperture control as that of the conventional imaging apparatus at the time of determining the follow light. Is obtained.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an imaging apparatus according to Embodiment 1 of the present invention.
FIG. 2 is a schematic diagram showing area division of the entire screen area in the area luminance detection unit of the imaging apparatus;
FIG. 3 is a configuration diagram showing weighting factors of a weighted 1 full-screen average calculating unit and a weighting 2 full-screen average calculating unit of the imaging apparatus;
FIG. 4 is a detection area diagram showing peak detection areas of a peak value 1 detection unit and a peak value 2 detection unit of the imaging apparatus;
FIG. 5 is a relational diagram showing the relationship between the luminance difference and the weight of the first area and the second area when the full screen average value is calculated by the full screen average value calculation unit of the imaging apparatus;
FIG. 6 is a relational diagram showing the relationship between the luminance difference between the first area and the second area and the weight when the weighted average error calculation unit of the imaging apparatus calculates the weighted average error.
FIG. 7 is a block diagram showing an imaging apparatus according to Embodiment 2 of the present invention.
FIG. 8 is a block diagram showing an imaging apparatus according to Embodiment 3 of the present invention.
FIG. 9 is a block diagram showing an imaging apparatus according to Embodiment 4 of the present invention.
FIG. 10 is a block diagram illustrating a conventional imaging device.
[Explanation of symbols]
101 Optical system
102 Aperture
103 Image sensor
104 ASP / A / D converter
105 Signal processing circuit
107 area luminance detector
108 Weighted 1 full screen average calculator
109 Weighted 2 full-screen average calculator
110 Peak value 1 detector
111 First / Second Region Luminance Comparison Unit
112 Average weight calculator
113 Whole screen average value calculator
114 Peak value error calculator
115 Average value error calculator
116 Peak weight calculator
117 Weighted average error calculator
118 Aperture control unit
701 Peak value 2 detector
702 Peak value calculator

Claims (6)

全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、
前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、
前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、
全画面領域内のピーク輝度値を検出するピーク値１検出部と、
前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、
前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、
前記ピーク値１検出部で得られたピーク輝度値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、
前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、
前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、
画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第 1 の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均値の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えた撮像装置。An area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;
A weighted 1 full-screen average calculating unit that calculates a weighted 1 full-screen average using each weight corresponding to each of the first areas in which the representative luminance value of each area and the weights are set to substantially equal weights ; ,
Weighted 2 full-screen average calculating unit for calculating weighted 2 full-screen average using each weight corresponding to each second area in which the representative luminance value and the weight of each area are set to the weight of the screen center weight When,
A peak value 1 detector for detecting a peak luminance value in the entire screen area;
A full screen average value calculation unit that calculates a full screen average value by weighted averaging the weighted 1 full screen average value and the weighted 2 full screen average value;
An average value error calculation unit for calculating an error (average value error) between the full screen average value obtained by the full screen average value calculation unit and a predetermined target value;
A peak value error calculating unit for calculating an error (peak value error) between the peak luminance value obtained by the peak value 1 detecting unit and a predetermined target value;
A weighted average error calculation unit for calculating a weighted average error the said peak value error obtained from the obtained average value error calculation unit mean value error between the peak value error calculator weighted average,
An aperture controller that determines the aperture of the aperture to follow the weighted average error to 0;
By subtracting the average brightness of the second area including the center of the screen from the average brightness of the first area including the upper part of the screen, a brightness difference Control (hereinafter abbreviated as CNT) between the first area and the second area is obtained. A first and second area luminance comparison unit that calculates a value, determines that the luminance difference CNT is positive if the value of the luminance difference CNT is positive, and determines the intensity of the backlight based on the magnitude of the value of the luminance difference CNT ;
When the value of the luminance difference CNT is negative from the calculation result of the first and second region luminance comparison units, the value of the luminance difference CNT is set so that the weight of the weighted 1 full screen average value becomes 100%. Is positive, that is, according to the intensity of the backlight, that is, according to the value of the luminance difference CNT, the weighting 2 full screen average value and the weighting are gradually increased so that the weighting of the weighting 2 full screen average value gradually increases. an average weight calculation unit that calculates a 2 weight weighted average of the entire screen average value,
From the calculation result of the first and second region luminance comparison units, when the value of the luminance difference CNT is negative, the weight ratio of the average value error and the peak value error is 3: 1. When the value of the brightness difference CNT is positive, the average value error and the peak value are increased according to the intensity of backlight, that is, the weight of the average value error gradually increases according to the value of the brightness difference CNT. an image pickup apparatus having a peak weight calculation unit that calculates the weight of the weighted average of the error. 全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、
前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、
前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、
全画面領域内のピーク１輝度値を検出するピーク値１検出部と、
画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、
前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、
前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、
前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、
前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、
前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、
前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、
画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えた撮像装置。An area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;
A weighted 1 full-screen average calculating unit that calculates a weighted 1 full-screen average using each weight corresponding to each of the first areas in which the representative luminance value of each area and the weights are set to substantially equal weights ; ,
Weighted 2 full-screen average calculating unit for calculating weighted 2 full-screen average using each weight corresponding to each second area in which the representative luminance value and the weight of each area are set to the weight of the screen center weight When,
A peak value 1 detector for detecting a peak 1 luminance value in the entire screen area;
A peak value 2 detector for detecting a peak 2 luminance value in the second area including the center of the screen;
A peak value calculator for calculating a peak value by weighted averaging the peak 1 luminance value and the peak 2 luminance value;
A full screen average value calculation unit that calculates a full screen average value by weighted averaging the weighted 1 full screen average value and the weighted 2 full screen average value;
An average value error calculation unit for calculating an error (average value error) between the full screen average value obtained by the full screen average value calculation unit and a predetermined target value;
A peak value error calculation unit that calculates an error (peak value error) between the peak value obtained by the peak value calculation unit and a predetermined target value;
A weighted average error calculation unit for calculating a weighted average error the said peak value error obtained from the obtained average value error calculation unit mean value error between the peak value error calculator weighted average,
An aperture controller that determines the aperture of the aperture to follow the weighted average error to 0;
By subtracting the average brightness of the second area including the center of the screen from the average brightness of the first area including the upper part of the screen, a brightness difference Control (hereinafter abbreviated as CNT) between the first area and the second area is obtained. A first and second area luminance comparison unit that calculates a value, determines that the luminance difference CNT is positive if the value of the luminance difference CNT is positive, and determines the intensity of the backlight based on the magnitude of the value of the luminance difference CNT ;
From the calculation result of the first and second area luminance comparison units, when the value of the luminance difference CNT is negative, the value of the luminance difference CNT is set so that the weight of the weighting 1 full screen average is 100%. In the positive case, the weighting 1 full-screen average value and the weighting 2 are set so that the weight of the weighting 2 full-screen average value gradually increases according to the intensity of backlight, that is, according to the value of the luminance difference CNT. an average weight calculation unit for calculating the weight of the weighted average of the entire screen average value,
From the calculation result of the first and second region luminance comparison units, when the value of the luminance difference CNT is negative, the weight ratio of the average value error and the peak value error is 3: 1. When the value of the brightness difference CNT is positive, the average value error and the peak value are increased according to the intensity of backlight, that is, the weight of the average value error gradually increases according to the value of the brightness difference CNT. an image pickup apparatus having a peak weight calculation unit that calculates the weight of the weighted average of the error. 全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、
前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、
全画面領域内のピーク１輝度値を検出するピーク値１検出部と、
画面中央を含む第２の領域のピーク２輝度値を検出するピーク値２検出部と、
前記ピーク１輝度値と前記ピーク２輝度値を加重平均しピーク値を算出するピーク値演算部と、
前記重み付け２全画面平均算出部で得られた重み付け２全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、
前記ピーク値演算部で得られたピーク値と所定の目標値との誤差（ピーク値誤差）を計算するピーク値誤差算出部と、
前記平均値誤差算出部から得られた前記平均値誤差と前記ピーク値誤差算出部から得られた前記ピーク値誤差を加重平均し加重平均誤差を算出する加重平均誤差算出部と、
前記加重平均誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、
画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記平均値誤差と前記ピーク値誤差の重みの比が３：１になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、輝度差分ＣＮＴの値の大きさに従い徐々に前記平均値誤差の重みが大きくなるように、前記平均値誤差と前記ピーク値誤差の加重平均の重みを算出するピーク重み算出部とを備えた撮像装置。An area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;
Weighted 2 full-screen average calculating unit for calculating weighted 2 full-screen average using each weight corresponding to each second area in which the representative luminance value and the weight of each area are set to the weight of the screen center weight When,
A peak value 1 detector for detecting a peak 1 luminance value in the entire screen area;
A peak value 2 detector for detecting a peak 2 luminance value in the second area including the center of the screen;
A peak value calculator for calculating a peak value by weighted averaging the peak 1 luminance value and the peak 2 luminance value;
An average value error calculating unit that calculates an error (average value error) between the weighted 2 full screen average value obtained by the weighted 2 full screen average calculating unit and a predetermined target value;
A peak value error calculation unit that calculates an error (peak value error) between the peak value obtained by the peak value calculation unit and a predetermined target value;
A weighted average error calculation unit for calculating a weighted average error the said peak value error obtained from the obtained average value error calculation unit mean value error between the peak value error calculator weighted average,
An aperture controller that determines the aperture of the aperture to follow the weighted average error to 0;
By subtracting the average brightness of the second area including the center of the screen from the average brightness of the first area including the upper part of the screen, a brightness difference Control (hereinafter abbreviated as CNT) between the first area and the second area is obtained. A first and second area luminance comparison unit that calculates a value, determines that the luminance difference CNT is positive if the value of the luminance difference CNT is positive, and determines the intensity of the backlight based on the magnitude of the value of the luminance difference CNT ;
From the calculation result of the first and second area luminance comparison units, when the value of the luminance difference CNT is negative, the weight ratio of the average value error to the peak value error is 3: 1. When the value of the luminance difference CNT is positive, the average value error and the peak value error are set so that the weight of the average value error gradually increases according to the intensity of the backlight, that is, according to the value of the luminance difference CNT. an image pickup apparatus having a peak weight calculation unit for calculating a weight of the weighted average. 全画面領域を複数の領域に分割し、各領域の代表輝度値を検出する領域輝度検出部と、
前記各領域の代表輝度値と各重みが略均等の重み付けに設定された第１の各領域に対応した各重みを用いて、重み付け１全画面平均値を算出する重み付け１全画面平均算出部と、
前記各領域の代表輝度値と各重みが画面中央重点の重み付けに設定された第２の各領域に対応した各重みを用いて、重み付け２全画面平均値を算出する重み付け２全画面平均算出部と、
前記重み付け１全画面平均値と前記重み付け２全画面平均値とを加重平均し全画面平均値を算出する全画面平均値演算部と、
前記全画面平均値演算部で得られた全画面平均値と所定の目標値との誤差（平均値誤差）を計算する平均値誤差算出部と、
前記平均値誤差を０へ追従させるべく絞り開度を決定する絞り制御部と、
画面上部を含む第１の領域の平均輝度から画面中央を含む第２の領域の平均輝度を減算して前記第１の領域と前記第２の領域の輝度差分Ｃｏｎｔｒｏｌ（以下、ＣＮＴと略す）の値を演算し、前記輝度差分ＣＮＴの値が正の場合は逆光と判断し、前記輝度差分ＣＮＴの値の大きさで逆光の強度を判断する第１・第２領域輝度比較部と、
前記第１・第２領域輝度比較部の演算結果から、前記輝度差分ＣＮＴの値が負の場合は、前記重み付け１全画面平均の重みが１００％になるように、前記輝度差分ＣＮＴの値が正の場合は、逆光の強度に従い、すなわち、前記輝度差分ＣＮＴの値の大きさに従い徐々に前記重み付け２全画面平均値の重みが大きくなるように、前記重み付け１全画面平均値と前記重み付け２全画面平均値の加重平均の重みを算出する平均重み算出部とを備えた撮像装置。An area luminance detection unit that divides the entire screen area into a plurality of areas and detects a representative luminance value of each area;
A weighted 1 full-screen average calculating unit that calculates a weighted 1 full-screen average using each weight corresponding to each of the first areas in which the representative luminance value of each area and the weights are set to substantially equal weights ; ,
Weighted 2 full-screen average calculating unit for calculating weighted 2 full-screen average using each weight corresponding to each second area in which the representative luminance value and the weight of each area are set to the weight of the screen center weight When,
A full screen average value calculation unit that calculates a full screen average value by weighted averaging the weighted 1 full screen average value and the weighted 2 full screen average value;
An average value error calculation unit for calculating an error (average value error) between the full screen average value obtained by the full screen average value calculation unit and a predetermined target value;
An aperture controller that determines the aperture of the aperture so that the average value error follows 0;
By subtracting the average brightness of the second area including the center of the screen from the average brightness of the first area including the upper part of the screen, a brightness difference Control (hereinafter abbreviated as CNT) between the first area and the second area is obtained. A first and second area luminance comparison unit that calculates a value, determines that the luminance difference CNT is positive if the value of the luminance difference CNT is positive, and determines the intensity of the backlight based on the magnitude of the value of the luminance difference CNT ;
If the value of the luminance difference CNT is negative from the calculation result of the first and second area luminance comparison units, the value of the luminance difference CNT is set so that the weight of the weighted 1 full screen average is 100%. In the positive case, the weighting 1 full-screen average value and the weighting 2 are set such that the weighting of the weighting 2 full-screen average value gradually increases according to the intensity of backlight, that is, according to the value of the luminance difference CNT. an image pickup apparatus having an average weight calculation unit that calculates the weight of the weighted average of the entire screen average value. 前記第２の各領域に対応した各重みは、画面中央重点の重み付けの設定に替えて、画面上部の領域に対して徐々に重みを小さく設定することを特徴とする請求項１乃至４のいずれかに記載の撮像装置。Each weight corresponding to the previous SL second each region, instead of the setting of the weighting of the screen center weighted, according to claim 1 to 4, characterized in that gradually smaller weight to a region of the top of the screen The imaging device according to any one of the above. 前記全画面領域を前記第１の領域と前記第２の領域をあわせた領域したことを特徴とする請求項１乃至５のいずれかに記載の撮像装置。Imaging device according to any one of claims 1 to 5, characterized in that the entire screen area and region combining the second region and the first region.
JP2000206465A 2000-07-07 2000-07-07 Imaging device Expired - Lifetime JP4013456B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2000206465A JP4013456B2 (en) 2000-07-07 2000-07-07 Imaging device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2000206465A JP4013456B2 (en) 2000-07-07 2000-07-07 Imaging device

Publications (2)

Publication Number Publication Date
JP2002023216A JP2002023216A (en) 2002-01-23
JP4013456B2 true JP4013456B2 (en) 2007-11-28

Family

ID=18703386

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2000206465A Expired - Lifetime JP4013456B2 (en) 2000-07-07 2000-07-07 Imaging device

Country Status (1)

Country Link
JP (1) JP4013456B2 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100557219B1 (en) * 2004-07-06 2006-03-07 삼성전자주식회사 Method and apparatus for compensating for automatic exposure
JP4755492B2 (en) * 2005-12-08 2011-08-24 パナソニック株式会社 Exposure amount control device and exposure amount control method
JP4832320B2 (en) * 2007-01-23 2011-12-07 富士フイルム株式会社 Imaging device
JP6046952B2 (en) * 2012-08-24 2016-12-21 キヤノン株式会社 Imaging apparatus and control method thereof

Also Published As

Publication number Publication date
JP2002023216A (en) 2002-01-23

Similar Documents

Publication Publication Date Title
US8154628B2 (en) Image processing apparatus and imaging apparatus and method
US7778478B2 (en) Image processing device, image processing program, image processing method, and electronic camera for controlling degree of noise reduction in image data
KR100557219B1 (en) Method and apparatus for compensating for automatic exposure
JP3412174B2 (en) Automatic exposure control device
US5353058A (en) Automatic exposure control apparatus
CN102724405B (en) Method and device for automatic exposure compensation of backlit scenes in video imaging system
US8773551B2 (en) Signal processing apparatus, solid-state image capturing apparatus, electronic information device, signal processing method, control program and storage medium
JPH10262181A (en) Exposure controller, exposure control method and camera
JP3478452B2 (en) Backlight detection method, backlight detection device, and imaging camera
JPH06292071A (en) Video camera
JP2899031B2 (en) Automatic exposure control device
JP4013456B2 (en) Imaging device
JP3838532B2 (en) White balance control method and apparatus
JP5591026B2 (en) Imaging apparatus and control method thereof
JP3092397B2 (en) Imaging device
JP4793001B2 (en) Image signal processing circuit and imaging system
JP3831934B2 (en) Camera automatic exposure device
US8154618B2 (en) Imaging apparatus and method for setting the same
JP3792555B2 (en) Brightness adjustment method and imaging apparatus
JPH01218178A (en) Image pickup device
JP2003299107A (en) Imaging device
JP3100815B2 (en) Camera white balance control method
JPH04340875A (en) Image pickup device
JP2001028760A (en) White balance control method and exposure control method for camera
JPH05207361A (en) Automatic exposure controller

Legal Events

Date Code Title Description
RD01 Notification of change of attorney

Free format text: JAPANESE INTERMEDIATE CODE: A7421

Effective date: 20050630

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20060131

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20070605

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20070731

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20070821

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20070903

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20100921

Year of fee payment: 3

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20100921

Year of fee payment: 3

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20110921

Year of fee payment: 4

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20120921

Year of fee payment: 5