JP3998840B2 - Projection display apparatus and image processing method therefor - Google Patents

Description

【００４１】
ここで、（Ｘｂ，Ｙｂ）は、図５に示すように、液晶パネル３００における正画像３４’の座標値を示している。（Ｈｎ，Ｖｎ）は、液晶パネル３００における歪み画像３４Ａ’の座標値を示している。Ｖｍは、液晶パネル３００の垂直方向の最大の座標値を示している。なお、これらの座標値は、図５に示す液晶パネル３００の左下点を座標原点ＯＲＧ（０，０）とした場合の座標値である。また、Ｌは、図１に示すように、投写型表示装置２０内の液晶パネル３００とスクリーン３０との距離を示している。θは、投写型表示装置２０から射出される光の光軸２０ｃとスクリーン３０の法線３０ｎとのずれ角を示している。数式１に示す関係については、例えば、特開平８−９８１１９号公報に記載されている。なお、本実施例においては、数式１に示す関係に基づいて、歪み画像３４Ａ’を生成するが、他の関係式を用いてもよい。
【００４２】
数式１の距離Ｌとずれ角θとの２つのパラメータは、図３に示すパラメータ測定部４０２あるいはパラメータ設定部４０４によって与えられる。パラメータ測定部４０２は、距離Ｌとずれ角θとを測定する機能を有している。距離Ｌは、例えば、カメラのオートフォーカスなどで用いられる距離計測用の超音波センサなどを投写型表示装置２０に備えることによって測定することができる。ずれ角θは、図１に示すように、あおり投写する際の投写型表示装置２０の傾きを調整する「足」の出方を測定することによって求めることができる。一方、パラメータ設定部４０４においては、ユーザが距離Ｌとずれ角θとの値をそれぞれ設定する。パラメータ測定部４０２あるいはパラメータ設定部４０４によって決定される２つのパラメータＬ，θを用いれば、数式１の関係から、歪み画像３４Ａ’内の各点に対応する正画像３４’内の点の位置関係を知ることができる。なお、パラメータ測定部４０２あるいはパラメータ設定部４０４によって決定された２つのパラメータＬ，θは、バスライン１００ａおよび図４の制御部２３０を介して読出アドレス生成部２６０に与えられる。
【００４３】
読出アドレス生成部２６０は、制御部２３０から入力されるパラメータＬ，θの値を用いて、数式１の関係に基づいて、前述のアドレス制御信号ＡＤＤを出力する。具体的には、図５に示す略台形の歪み画像３４Ａ’内に含まれる各画素位置（Ｈｎ，Ｖｎ）に対応する正画像３４’の画素位置（Ｘｂ，Ｙｂ）を特定するためのアドレス値をアドレス制御信号ＡＤＤとして出力する。また、図５の斜線部に示す略台形の歪み画像３４Ａ’内に含まれない画素位置に対応する正画像３４’の画素位置、液晶パネル３００の領域外となる。この場合には、アドレス制御信号ＡＤＤとして、対応する正画像３４’の画素が存在しないことを示すアドレス値（例えば、「０」）を出力する。アドレス制御信号ＡＤＤは、制御部２３０から供給される制御信号ＣＴＲＲに基づくタイミングで順次出力される。
【００４４】
上記のように、読出制御部２５０は、２つのフレームメモリ１４０，１４２に記憶された画像データＤＴ，ＤＴ’に含まれる複数の画素のうち、アドレス制御信号ＡＤＤによって指定されたアドレスに記憶された画素データのみを順次抽出することによって、歪み画像データＤＤＶを出力する。このとき、対応する正画像３４’の画素が存在しないことを示すアドレス値（例えば、「０」）が入力される場合には、その画素については、例えば、「黒」を示す画素データを出力する。これによって、読出制御部２５０は、図５に示す略台形の歪み画像３４Ａ’を含む矩形の画像を表す歪み画像データＤＤＶを出力することができる。
【００４５】
輝度調整部２８０は、読出制御部２５０から出力された歪み画像データＤＤＶによって表される歪み画像の輝度を調整して、調整済み画像データＡＤＶを出力する機能を有している。
【００４６】
図６は、輝度調整部２８０における処理を示す説明図である。図６（Ａ）は、輝度調整部２８０に入力される歪み画像データＤＤＶによって表される略台形の歪み画像が液晶パネル３００にそのまま形成された場合を示している。図６（Ｂ）〜（Ｄ）は、輝度調整部２８０から出力される調整済み画像データＡＤＶ１〜ＡＤＶ３によって表される３種類の略台形の歪み画像が液晶パネル３００に形成された場合を示している。なお、図６（Ａ）〜（Ｄ）に示す画像は、図４のフレームメモリ１４０，１４２内に全て同じ画素値を有する単色の「べた画像」を表す画像データＤＴ，ＤＴ’が記憶されている場合に出力される歪み画像データＤＤＶおよび調整済み画像データＡＤＶ１〜ＡＤＶ３によって表される画像を示している。したがって、液晶パネル３００に形成される図６（Ａ）の歪み画像の画像内の輝度はほぼ均一であるが、図６（Ｂ）〜（Ｄ）の調整済み画像の画像内の輝度は調整されている。
【００４７】
輝度調整部２８０における図６（Ｂ）〜（Ｄ）に示す処理は、図３の輝度調整モード設定部４０６において設定される輝度調整モードに従って行われる。ユーザにより設定された輝度調整モードは、図４に示すようにバスライン１００ａおよび制御部２３０を介して、モード情報信号ＭＤとして輝度調整部２８０に入力される。
【００４８】
図６（Ｂ）は、第１の輝度調整モードを用いた場合の調整済み画像データＡＤＶ１によって表される調整済み画像が、液晶パネル３００に形成された場合を示している。第１の輝度調整モードを用いる場合には、略台形の歪み画像の上底の水平ラインが基準ラインとして設定され、各水平ラインの輝度が調整される。ここで、「基準ライン」とは、輝度調整の基準となる水平ラインを意味し、他の水平ラインは基準ラインとの関係により輝度が調整される。本実施例の輝度調整においては、各水平ラインに含まれる複数の画素のそれぞれに対して、水平ライン毎に変化する第１の係数Ｋ１（ｔ）が乗じられる。第１の係数Ｋ１（ｔ）は、次の数式２によって表される。
【００４９】
【数２】

【００５０】
ここで、図６（Ａ）に示すように、ｔは、略台形の歪み画像を構成する各水平ラインを下底から数えたときのｔ番目の水平ラインを示しており、ｔmax は、略台形の歪み画像の上底の水平ラインであるｔmax 番目の水平ラインを示す。Ｈ（ｔ）は、略台形の歪み画像を構成するｔ番目の水平ラインの幅を意味し、Ｖ（ｔ）は、ｔ番目の水平ラインの高さを意味する。したがって、Ｈ（ｔmax ）は、略台形の歪み画像のｔmax 番目の水平ライン（上底の水平ライン）の幅を示しており、Ｖ（ｔmax ）は、上底の水平ラインの高さを示している。また、Ｈｍは、図５に示すＨｍと同じであり、０番目の水平ライン（下底の水平ライン）の幅Ｈ（０）を意味する。また、Ｓ１は、０より大きな定数である。なお、定数Ｓ１については、さらに、後述する。
【００５１】
数式２から分かるように、第１の係数Ｋ１（ｔ）は、基準ラインであるｔmax 番目の上底の水平ラインについては「Ｓ１」となり、０番目の下底の水平ラインについては「Ｓ１・（Ｈ（ｔmax ）・Ｖ（ｔmax ））／（Ｈｍ・Ｖｍ）」となるように変化する。具体的には、数式２のＨｍが含まれる第１の括弧内の関数は、上底ラインであるｔmax 番目の水平ラインにおいて「１」となり、下底ラインである０番目の水平ラインにおいて「Ｈ（ｔmax ）／Ｈｍ」となるように、水平ラインｔに応じて線形に変化する。また、Ｖｍが含まれる第２の括弧内の関数は、上底ラインであるｔmax 番目の水平ラインにおいて「１」となり、下底ラインである０番目の水平ラインにおいて「Ｖ（ｔmax ）／Ｖｍ」となるように、水平ラインｔに応じて線形に変化する。したがって、第１の係数Ｋ１（ｔ）は、略台形の歪み画像を構成する上底側の水平ラインから下底側の水平ラインに向かうにつれて単調に小さくなる。
【００５２】
なお、上記の説明から分かるように、数式２のＨｍが含まれる第１の括弧内の関数は、略台形の歪み画像の水平方向の歪みの補正量に基づく水平方向に関する輝度補正量を意味する。また、数式２のＶｍが含まれる第２の括弧内の関数は、略台形の歪み画像の垂直方向の歪みの補正量に基づく垂直方向に関する輝度補正量を意味する。
【００５３】
このとき、液晶パネル３００に形成される略台形の歪み画像の輝度は、図６（Ｂ）に示すように調整される。すなわち、液晶パネル３００に表示される基準ラインである上底の水平ラインのライン画像の輝度についてはＳ１倍に調整され、他の水平ラインのライン画像の輝度については、上底側の水平ラインから下底側の水平ラインに向かうにつれて単調に小さくなる。なお、図６（Ｂ）においては、図示の便宜上、略台形の歪み画像の輝度変化は６つの領域に区分して表されているが、実際には、各水平ライン毎に単調に輝度が変化している。このようにして、液晶パネル３００に形成される略台形の歪み画像の輝度を調整した調整済み画像データＡＤＶ１を生成することにより、スクリーン３０上に表示される正画像の輝度を、その画像全体でほぼ均一にすることが可能となる。
【００５４】
なお、スクリーン３０上に表示される画像が比較的暗い場合には、定数Ｓ１を１より大きな値に設定することによって、画像全体を明るくすることができる。すなわち、第１の係数Ｋ１（ｔ）に含まれる定数Ｓ１は、スクリーン３０上に表示される正画像の画像全体の輝度をほぼ均一に保ったまま、さらに、画像全体の輝度を調整するための乗数である。なお、定数Ｓ１の設定は、例えば、輝度を調整する前の略台形の歪み画像に含まれる全ての画素の平均の画素値を求めることによってうまく行うことができる。この平均の画素値は、読出制御部２５０（図４）において略台形の歪み画像を構成する各画素データを読み出す際に求めることができる。求められた平均の画素値が比較的小さい場合には、スクリーン３０上に表示される画像を明るくするために定数Ｓ１として１を超える比較的大きな値を設定すればよい。
【００５５】
図６（Ｃ）は、第２の輝度調整モードを用いた場合の調整済み画像データＡＤＶ２によって表される調整済み画像が、液晶パネル３００に形成された場合を示している。第２の輝度調整モードを用いる場合には、略台形の歪み画像の下底の水平ラインが基準ラインとして設定され、各水平ラインの輝度が調整される。この場合には、各水平ラインに含まれる複数の画素のそれぞれに対して、水平ライン毎に変化する第２の係数Ｋ２（ｔ）が乗じられる。第２の係数Ｋ２（ｔ）は、次の数式３によって表される。
【００５６】
【数３】

【００５７】
数式３から分かるように、第２の係数Ｋ２（ｔ）は、基準ラインである０番目の下底の水平ラインについて「Ｓ２」となり、ｔmax 番目の上底の水平ラインについて「Ｓ２・（Ｈｍ・Ｖｍ）／（Ｈ（ｔmax ）・Ｖ（ｔmax ））」となるようい変化する。具体的には、数式３のＨｍが含まれる第１の括弧内の関数は、上底ラインであるｔmax 番目の水平ラインにおいて「Ｈｍ／Ｈ（ｔmax ）」となり、下底ラインである０番目の水平ラインにおいて「１」となるように、水平ラインｔに応じて線形に変化する。また、Ｖｍが含まれる第２の括弧内の関数は、上底ラインであるｔmax 番目の水平ラインにおいて「Ｖｍ／Ｖ（ｔmax ）」となり、下底ラインである０番目の水平ラインにおいて「１」となるように、水平ラインｔに応じて線形に変化する。したがって、この場合にも、第２の係数Ｋ２（ｔ）は、略台形の歪み画像を構成する上底側の水平ラインから下底側の水平ラインに向かうにつれて単調に小さくなっている。
【００５８】
このとき、液晶パネル３００に形成される略台形の歪み画像の輝度は、図６（Ｃ）に示すように、基準ラインである下底の水平ラインのライン画像の輝度についてはＳ２倍に調整され、他の水平ラインのライン画像の輝度については、下底側の水平ラインから上底側の水平ラインに向かうにつれて単調に大きくなる。第２の係数Ｋ２（ｔ）を用いる場合にも、第１の係数Ｋ１（ｔ）を用いる場合と同様に、スクリーン３０上に表示される正画像の輝度を、その画像全体においてほぼ均一にすることができる。
【００５９】
ただし、数式３に示す第２の係数Ｋ２（ｔ）は、定数Ｓ２が「１」である場合には常に１以上の値となるため、略台形の歪み画像の上底側のライン画像に含まれる画素の画素値が大きな値を示す場合には、第２の係数Ｋ２（ｔ）を乗じることによって画素値のダイナミックレンジを超えてしまう場合がある。例えば、各画素の画素値が８ビットで構成されている場合には、第２の係数Ｋ２（ｔ）を乗じることによって、ダイナミックレンジ（０〜２５５）を超える値となってしまう場合がある。このような場合には、第２の係数Ｋ２（ｔ）に含まれる定数Ｓ２を１より小さな値に設定することによって、画像全体を暗くすることが好ましい。なお、定数Ｓ２の設定は、上記の定数Ｓ１の設定と同様に、輝度を調整する前の略台形の歪み画像に含まれる全ての画素の平均の画素値に基づいて行うことができる。
【００６０】
図６（Ｄ）は、第３の輝度調整モードを用いた場合の調整済み画像データＡＤＶ３によって表される調整済み画像が、液晶パネル３００に形成された場合を示している。第３の輝度調整モードを用いる場合には、略台形の歪み画像の各水平ラインに含まれる複数の画素の画素値の平均が最大となる水平ラインが基準ラインとして設定され、各水平ラインの輝度が調整される。輝度調整においては、各水平ラインに含まれる複数の画素のそれぞれに対して、水平ライン毎に変化する第３の係数Ｋ３（ｔ）が乗じられる。第３の係数Ｋ３（ｔ）は、次の数式４によって表される。
【００６１】
【数４】

【００６２】
ここで、ｔｌは、基準ラインとなるｔｌ番目の水平ラインを示している。また、Ｈ（ｔｌ）は、略台形の歪み画像を構成するｔｌ番目の水平ラインの幅を示している。
【００６３】
数式４から分かるように、第３の係数Ｋ３（ｔ）は、略台形の歪み画像を構成する各水平ラインについては、上底側の水平ラインから下底側の水平ラインに向かうにつれて単調に小さくなり、また、基準ラインｔｌについては「Ｓ３」となる。
【００６４】
このとき、液晶パネル３００に形成される略台形の歪み画像の輝度は、図６（Ｄ）に示すように調整される。液晶パネル３００に表示される基準ラインであるｔｌ番目の水平ラインのライン画像の輝度についてはＳ３倍に調整され、他の水平ラインのライン画像の輝度については、上底側の水平ラインから下底側の水平ラインに向かうにつれて単調に小さくなる。したがって、第３の係数Ｋ３（ｔ）を用いる場合にも、上記の第１および第２の係数Ｋ１（ｔ），Ｋ２（ｔ）を用いる場合と同様に、スクリーン３０上に表示される正画像の輝度を、その画像全体においてほぼ均一にすることができる。また、この場合には、略台形の歪み画像の各水平ラインに含まれる複数の画素の画素値の平均が最大となる水平ラインが基準ラインとして設定されているので、下底の水平ラインを基準ラインとする第２の輝度調整モード（数式３）よりも、各水平ラインに含まれる画素値がダイナミックレンジを超える恐れが少なくなるという利点がある
【００６５】
なお、上記においては、略台形の歪み画像の各水平ラインに含まれる複数の画素の画素値の平均が最大となる水平ラインが基準ラインとして設定されているが、略台形の歪み画像に含まれる全ての画素うち画素値が最大となる画素を含む水平ラインを基準ラインとして設定してもよい。
【００６６】
このように、輝度調整部２８０（図４）は、制御部２３０から入力されるモード情報信号ＭＤに従って、輝度調整モードを決定する。また、読出アドレス生成部２６０から入力される種々の水平ライン（上底ライン、下底ライン、基準ラインなど）の幅や高さなどの係数決定情報ＫＤに基づいて、各輝度調整モードにおいて必要な輝度調整係数Ｋ１（ｔ）〜Ｋ３（ｔ）の内容を決定する。そして、決定された輝度調整係数に基づいて、歪み画像データＤＤＶを構成する各ライン画像信号の画素値を調整し、調整済み画像データＡＤＶを出力する。
【００６７】
ところで、上記の第１ないし第３の係数Ｋ１（ｔ）〜Ｋ３（ｔ）では、上底の水平ラインから下底の水平ラインに向かって、水平方向（ライン方向）についてのみならず垂直方向（高さ方向）についても、その輝度補正量は線形で変化している。例えば、数式２に示す第１の係数Ｋ１（ｔ）では、垂直方向の輝度補正量を示すＶｍが含まれる第２の括弧内の関数は、水平ラインの位置ｔに応じて線形に変化する関数である。しかし、実際には、略台形の歪み画像は、数式１に基づいて、図５に示すように補正される。すなわち、略台形の歪み画像は、水平方向については、上底の水平ラインから下底の水平ラインまでほぼ線形に補正されているが、垂直方向については、上底側ほど大きな割合で補正されている。したがって、垂直方向の輝度補正量については、数式１に忠実に従った補正量を用いることが望ましい。こうすれば、歪み画像を生成する際の補正量と対応した輝度調整を行うことができるので、スクリーン上に表示される画像内の輝度分布をよりかなり均一にすることができるという利点がある。ただし、上記のように垂直方向について線形に変化する輝度補正量を用いる場合には、スクリーン３０上に表示される正画像の画像内の輝度を容易にほぼ均一にすることができるという利点がある。
【００６８】
また、上記の係数Ｋ１（ｔ）〜Ｋ３（ｔ）は、略台形の歪み画像を構成する各ラインの水平方向の輝度補正量にのみ基づいて決定されていてもよい。この場合には、第１の係数Ｋ１（ｔ）については、例えば、数式２の水平方向の輝度補正量に関するＨｍが含まれる第１の括弧内の関数を２乗したものを用いればよい。こうしても、スクリーン３０上に表示される正画像の輝度分布をほぼ均一にすることが可能となる。
【００６９】
また、上記の第１ないし第３の係数Ｋ１（ｔ）〜Ｋ３（ｔ）に代えて、次の数式５で表される係数Ｋ（ｔ）を用いてもよい。
【００７０】
【数５】

【００７１】
ここで、ｓは、０より大きな定数である。
【００７２】
上記の数式５を用いても、スクリーン３０上に表示される正画像の輝度分布をほぼ均一にすることができる可能性がある。
【００７３】
上記の説明から分かるように、輝度調整係数は、略矩形の歪み画像を表す歪み画像データＤＤＶを生成する際に用いられる歪みの補正量に基づいて決定されていればよい。
【００７４】
図３の駆動部１５０は、供給された調整済み画像データＡＤＶに従って液晶パネル３００を駆動する。液晶パネル３００には、光源部３１０から射出される光が照射されており、液晶パネル３００が駆動されることにより液晶パネル３００に調整済み画像データＡＤＶによって表される輝度が調整された歪み画像が形成される。液晶パネル３００に形成された歪み画像は、投写光学系によって投写され、スクリーン３０上に輝度がほぼ均一に調整された正画像が表示される。
【００７５】
以上、説明したように、本実施例においては、液晶パネル３００に形成される略台形の歪み画像を構成する各ラインのライン画像の輝度を調整する。これにより、スクリーン３０上において、歪みのない正画像を表示させることができるとともに、表示される正画像の画像内の輝度をうまく調整することが可能となる。
【００７６】
なお、この発明は上記の実施例や実施形態に限られるものではなく、その要旨を逸脱しない範囲において種々の態様において実施することが可能であり、例えば以下のような変形も可能である。
【００７７】
（１）上記実施例では、画像形成部として、液晶パネル３００を用いているが、これに代えてマイクロミラー型光変調装置を用いるようにしてもよい。マイクロミラー型光変調装置としては、例えば、ＤＭＤ（デジタル・マイクロミラー・デバイス）（ＴＩ社の登録商標）を用いることができる。また、画像形成部として、高輝度ブラウン管を用いることも可能である。本発明の画像形成部としては、一般に、画像信号を画像光に変換する種々の装置を利用することができる。
【００７８】
なお、画像形成部として、上記の高輝度ブラウン管を用いる場合には、通常、最高輝度の状態は利用されていないので、略台形の歪み画像の上底側のライン画像の輝度を、対応する矩形の正画像のライン画像の輝度よりも大きくすることが容易である。一方、液晶パネルにおいては、通常、パネルのほぼ最高の光透過率まで利用されているため、略台形の歪み画像の上底側のライン画像の輝度を、対応する矩形の正画像のライン画像の輝度と同等にして、略台形の歪み画像の下底側のライン画像の輝度を小さくすることが好ましい。
【００７９】
（２）上記実施例では、図４に示す読出制御部２５０は、略台形の歪み画像を構成するのに必要な画素データをフレームメモリ１４０，１４２からそのまま読み出すことにより歪み画像データＤＤＶを出力しているが、略台形の歪み画像データＤＤＶに含まれる各画素データは補完されていてもよい。画素データの補完は、例えば、略台形の歪み画像において隣接する画素の画素データ間で行えばよい。
【００８０】
（３）上記実施例では、図１（Ｂ）に示すように、投写型表示装置２０から射出される画像を上方にあおり投写する場合について説明したが、投写型表示装置２０を天井などに吊り下げて画像を下方にあおり投写する場合についても、同様に本発明を適用することができる。この場合には、液晶パネル３００上に、図６に示す略台形の歪み画像と上下逆向きの画像を形成すればよい。この場合にも、略台形の歪み画像の輝度は、上底（短辺）の水平ラインから下底（長辺）の水平ラインに向かって単調に小さくなるように調整すればよい。
【００８１】
（４）上記実施例では、図４に示すように、本発明の画像歪み調整部と輝度調整部とに相当する読出制御部２５０と読出アドレス生成部２６０と輝度調整部２８０とが、投写型表示装置２０の内部に備えられている場合について説明したが、他の画像処理装置が、画像歪み調整部と輝度調整部とを備えるようにしてもよい。この場合には、他の画像処理装置が、入力される画像信号を調整することによって調整済み画像信号を生成し、画像形成部と投写光学系とを備える投写部に供給すればよい。なお、この場合には、他の画像処理装置と投写部とが、本発明の投写型表示装置に相当する。
【図面の簡単な説明】
【図１】投写型表示装置を用いて投写される画像を示す説明図。
【図２】あおり投写する場合に、スクリーン３０上に表示される画像と液晶パネル３００に形成される画像との関係を示す説明図。
【図３】本発明の実施例としての投写型表示装置の電気的構成を主として示すブロック図。
【図４】ビデオプロセッサ２００の内部構成の一例を示す説明図。
【図５】液晶パネル３００に形成される正画像３４’と歪み画像３４Ａ’との関係を示す説明図。
【図６】輝度調整部２８０における処理を示す説明図。
【符号の説明】
２０…投写型表示装置
３０…スクリーン
１００…ＣＰＵ
１００ａ…バスライン
１１０…ビデオデコーダ
１２０…同期分離回路
１２２…ＡＤ変換部
１４０，１４２…フレームメモリ
１５０…駆動部
２００…ビデオプロセッサ
２１０…サンプリングクロック生成部
２２０…データセレクタ
２３０…制御部
２４０…書込制御部
２５０…読出制御部
２６０…読出アドレス生成部
２８０…輝度調整部
３００…液晶パネル
３１０…光源部
３２０…投写レンズ
４０２…パラメータ測定部
４０４…パラメータ設定部
４０６…輝度調整モード設定部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for adjusting distortion of an image generated when tilting projection in a projection display device.
[0002]
[Prior art]
FIG. 1 is an explanatory diagram showing an image projected using a projection display device. 1A shows a first positional relationship between the projection display device 20 and the screen 30, and FIG. 1B shows a display on the screen 30 in the positional relationship shown in FIG. An image 32 is shown. In the positional relationship shown in FIG. 1A, the optical axis 20c of light representing an image emitted from the projection display device 20 (hereinafter also referred to as “image light”) and the normal line 30n of the screen 30 match. Yes. In this case, as shown in FIG. 1B, an image 32 displayed on the screen 30 is an image without distortion (hereinafter, an image without distortion is also referred to as a “normal image”).
[0003]
On the other hand, FIG. 1C shows a second positional relationship between the projection display device 20 and the screen 30, and FIG. 1D shows the position on the screen 30 in the positional relationship shown in FIG. An image 34 displayed in FIG. In the positional relationship shown in FIG. 1C, the optical axis 20c of the image light emitted from the projection display device 20 and the normal line 30n of the screen 30 do not coincide with each other, and are shifted by an angle θ. In such a case, as shown in FIG. 1D, the image 34 displayed on the screen 30 is a substantially trapezoidally distorted image (hereinafter, the distorted image is also referred to as a “distorted image”).
[0004]
As shown in FIG. 1C, when the image light is tilted and projected, for example, an image distorted in a substantially trapezoid shape is formed in advance on a liquid crystal panel or the like provided in the projection display device 20. Thus, the image projected on the screen 30 is displayed as a normal image 34A having no distortion as shown by the broken line in FIG. Note that the normal image 34A in FIG. 1D is substantially the same as the normal image 32 in FIG.
[0005]
FIG. 2 is an explanatory diagram showing a relationship between an image displayed on the screen 30 and an image formed on the liquid crystal panel 300 when tilting projection is performed. 2A shows an image displayed on the screen 30, and FIG. 2B shows an image formed on the liquid crystal panel 300. In practice, the image displayed on the screen 30 is sufficiently larger than the image formed on the liquid crystal panel 300, but in FIG. 2, they are drawn with approximately the same size for the convenience of illustration. In the liquid crystal panel 300 of FIG. 2B, the image displayed when the normal image 34 ′ formed on the entire display area of the panel is projected is a distorted image 34 shown in FIG. On the other hand, in the liquid crystal panel 300 of FIG. 2B, an image displayed when the substantially trapezoidal distortion image 34A ′ formed in the display area of the panel is projected as shown in FIG. 2A. A normal image 34A can be obtained.
[Patent Document 1]
JP 2000-10185 A
[0006]
[Problems to be solved by the invention]
However, as shown in FIG. 2, when a substantially trapezoidal distortion image 34 </ b> A ′ is formed on the liquid crystal panel 300, there is a problem that the luminance of the normal image 34 </ b> A displayed on the screen 30 changes in the image. there were. That is, the upper base 34A′u (FIG. 2B) of the substantially trapezoidal distortion image 34A ′ of the liquid crystal panel 300 is the upper base 34 of the normal image 34 on the screen 30.AThe enlargement ratio when enlarged to u (FIG. 2A) is that the lower base 34A'd (FIG. 2B) of the substantially trapezoidal distortion image 34A 'of the liquid crystal panel 300 isunderBottom 34AdIt is larger than the enlargement ratio in the case of enlargement (FIG. 2A). Therefore, for example, when the distorted image 34A ′ formed on the liquid crystal panel 300 is a single color “solid image” having the same luminance in the image, the upper base of the normal image 34A formed on the screen 30 is displayed. The luminance at 34Au is smaller than the luminance at the lower base 34Ad and becomes darker.
[0007]
The present invention has been made to solve the above-described problems in the prior art, and in a projection display device, when projecting an image, the brightness in the projected image can be adjusted well. The purpose is to provide technology that can be used.
[0008]
[Means for solving the problems and their functions and effects]
In order to solve at least a part of the problems described above, an apparatus of the present invention is a projection display apparatus,
An image adjustment unit for adjusting an input image signal and outputting an adjusted image signal;
An image forming unit for forming an image based on the adjusted image signal;
A projection optical system for projecting an original image formed on the image forming unit onto a screen;
With
The image adjustment unit
In order to correct image distortion that occurs when a substantially rectangular original image is projected on the screen, the input image signal is adjusted so that a substantially trapezoidal original image is formed in the image forming unit. An image distortion adjustment unit;
In order to correct a change in luminance in an image that occurs when the substantially trapezoidal original image formed in the image forming unit is projected on the screen, a plurality of elements constituting the substantially trapezoidal original imageNo laThe brightness of the in-image monotonically decreases from the upper bottom line image to the lower bottom line image of the substantially trapezoidal original image.The compoundA brightness adjustment unit for adjusting each of the number of line images;
It is characterized by providing.
[0009]
By using the apparatus of the present invention, when the substantially trapezoidal distortion image formed in the image forming unit is projected, the image without distortion is displayed on the screen, and the image in the image without distortion is displayed. The brightness can be adjusted well.
[0010]
In the present specification, of the two substantially parallel sides forming the substantially trapezoidal shape, the short side is referred to as “upper base” and the long side is referred to as “lower base”.
[0011]
In the above apparatus,
The luminance adjustment unit uses the upper-bottom line image among the plurality of line images included in the substantially trapezoidal original image as a reference line image, and each of the luminance adjustment coefficients decreases monotonously as it goes to the lower-bottom line image. It is preferable to adjust the luminance by multiplying the pixel values of a plurality of pixels included in the line image.
[0012]
In this way, if the line image on the upper base of the substantially trapezoidal original image is used as the reference line image, each line image on the lower base side can be multiplied by a brightness correction coefficient smaller than that of the reference line image. It becomes possible to adjust the brightness in the image of the image displayed above well.
[0013]
In the above apparatus,
The luminance adjustment unit uses a lower-bottom line image among the plurality of line images included in the substantially trapezoidal original image as a reference line image, and sets a luminance adjustment coefficient that increases monotonously toward the upper-bottom line image. The luminance may be adjusted by multiplying the pixel values of a plurality of pixels included in the line image.
[0014]
In this way, when the line image at the lower base of the substantially trapezoidal original image is used as the reference line image, the screen image is obtained by multiplying each line image on the upper base side by a luminance correction coefficient larger than that of the reference line image. The brightness of the image displayed above can be adjusted well.
[0015]
In the above apparatus,
The brightness adjustment unit uses, as a reference line image, a line image in which an average of pixel values of a plurality of pixels included in each line image among the plurality of line images included in the substantially trapezoidal original image is the maximum. The brightness is adjusted by multiplying the pixel value of a plurality of pixels included in each line image by a brightness adjustment coefficient that increases monotonously as it goes to the upper bottom line image from the reference line image, and the lower bottom line image from the reference line is adjusted. The brightness may be adjusted by multiplying the pixel value of a plurality of pixels included in each line image by a brightness adjustment coefficient that decreases monotonously as it goes.
[0016]
In this way, the line image having the highest average luminance of each pixel in the substantially trapezoidal original image can be used as the reference line image. Based on this reference line imageTaikiIf the degree adjustment coefficient is multiplied by the pixel values of a plurality of pixels included in each line image, the luminance of the image displayed on the screen can be adjusted well.
In the above apparatus,
The brightness adjusting unit is
Among the plurality of line images included in the substantially trapezoidal original image, a plurality of brightness adjustment coefficients that monotonously decrease toward the lower bottom line image are included in each line image, with the upper bottom line image as a reference line image. A first luminance adjustment mode for adjusting the luminance by multiplying the pixel value of the pixel;
Among the plurality of line images included in the substantially trapezoidal original image, the line image at the lower base is used as a reference line image, and a plurality of luminance adjustment coefficients that are monotonically increased toward the line image at the upper base are included in each line image. A second luminance adjustment mode for adjusting the luminance by multiplying the pixel value of the pixel;
Among the plurality of line images included in the substantially trapezoidal original image, a line image having an average pixel value of a plurality of pixels included in each line image is defined as a reference line image, and an upper base from the reference line image. The brightness is adjusted by multiplying the pixel value of a plurality of pixels included in each line image by a brightness adjustment coefficient that increases monotonously as it goes to the line image, and decreases monotonously as it goes to the line image below the reference line. A third brightness adjustment mode for adjusting the brightness by multiplying the brightness adjustment coefficient by the pixel values of a plurality of pixels included in each line image;
Have
The brightness adjustment unit may perform brightness adjustment according to a brightness adjustment mode selected from the three brightness adjustment modes.
[0017]
In the above apparatus,
The luminance adjustment coefficient is preferably determined based on a distortion correction amount used when adjusting the input image signal in the image distortion adjustment unit.
[0018]
In this way, since the pixel values of a plurality of pixels included in each line image can be adjusted based on the distortion correction amount used in the image distortion adjustment unit, the luminance adjustment corresponding to the distortion correction amount is performed. Can be performed.
[0019]
In the above apparatus,
The brightness adjustment coefficient is, Determined using a correction amount of distortion in the height direction of the substantially trapezoidal original imageIt is preferable.
[0020]
In this way, the brightness of the image displayed on the screen can be adjusted well.
[0021]
In the above apparatus,
The brightness adjustment coefficient is further, Determined using a correction amount of distortion in the line direction of the substantially trapezoidal original imageIt is preferable.
[0022]
In this way, it is possible to more accurately determine the luminance adjustment coefficient that can adjust the luminance of the image displayed on the screen.
[0023]
In the present specification, the “height direction” means a direction perpendicular to two substantially parallel sides (upper base and lower base) constituting a substantially trapezoid.
[0024]
In the above apparatus,
The brightness adjustment coefficient is
From the upper-bottom line image determined based on the width of the upper-bottom line image of the substantially trapezoidal original image formed in the image forming unit and the width of the lower-bottom line image of the substantially rectangular original image The first adjustment coefficient that changes linearly toward the bottom-bottom line image and the luminance correction amount in the height direction of the luminance adjustment coefficient are the height of the substantially trapezoidal original image formed in the image forming unit. And a second adjustment coefficient that linearly changes from the upper base line image to the lower base line image, which is determined based on the height of the substantially rectangular original image, There may be.
[0025]
Even in this case, the brightness of the image displayed on the screen can be adjusted relatively well.
In the above apparatus,
The luminance adjustment coefficient may include a multiplier for adjusting the luminance in the image displayed on the screen over the entire image.
[0026]
A method of the present invention is an image processing method for processing an original image formed in an image forming unit of a projection display device that projects and displays an image on a screen,
A step of adjusting an input image signal so that a substantially trapezoidal original image is formed in the image forming unit in order to correct image distortion caused when a substantially rectangular original image is projected on the screen; When,
In order to correct a change in luminance in an image that occurs when the substantially trapezoidal original image formed in the image forming unit is projected on the screen, a plurality of elements constituting the substantially trapezoidal original imageNo laThe brightness of the in-image monotonically decreases from the upper bottom line image to the lower bottom line image of the substantially trapezoidal original image.The compoundAdjusting each of the number of line images;
It is characterized by providing.
[0027]
Even when the method of the present invention is used, the same operation and effect as the above-described apparatus are provided, and when a substantially trapezoidal distortion image formed in the image forming unit is projected, an image without distortion is displayed on the screen. In addition, it is possible to adjust the luminance in the image of the displayed image without distortion.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
A. Projection display:
Next, embodiments of the present invention will be described based on examples. FIG. 3 is a block diagram mainly showing an electrical configuration of a projection display apparatus as an embodiment of the present invention. This apparatus includes a CPU 100, a video decoder 110, a synchronization separation circuit 120, an AD conversion unit 122, a video processor 200, two sets of frame memories 140 and 142, a drive unit 150, and a liquid crystal panel 300. Is provided. The CPU 100 and the video processor 200 are connected via a bus line 100a. In addition, a parameter measuring unit 402, a parameter setting unit 404, and a luminance adjustment mode setting unit 406 for generating an image formed on the liquid crystal panel 300 are connected to the bus line 100a. Further, this apparatus includes a light source unit 310 and a projection lens 320.
[0029]
3 corresponds to the image forming unit of the present invention, and the light source unit 310 and the projection lens 320 correspond to the projection optical system of the present invention. Further, each part excluding the liquid crystal panel 300, the light source part 310, and the projection lens 320 shown in FIG. 3 corresponds to the image adjustment part of the present invention.
[0030]
The video decoder 110 receives an analog image signal AV1 output from a video recorder or a television. The analog image signal AV1 is a signal in which a luminance signal, a color signal, and a synchronization signal are superimposed. The video decoder 110 generates and outputs a digital image signal DV1 composed of digital color signals of R, G, B from the input analog image signal AV1, and also outputs vertical and horizontal synchronization signals VSYNC1, HSYNC1. Are output separately. Note that the output digital image signal DV1 and the two synchronization signals VSYNC1 and HSYNC1 are input to the video processor 200 described later.
[0031]
On the other hand, the analog image signal AV2 output from the personal computer is input to the synchronization separation circuit 120. The analog image signal AV2 includes an analog color signal and vertical and horizontal synchronization signals. The synchronization separation circuit 120 separates and outputs the vertical and horizontal synchronization signals VSYNC2 and HSYNC2 and the analog color signal AV2 'composed of three color signals from the analog image signal AV2. The analog color signal AV <b> 2 ′ is input to the AD conversion unit 122.
[0032]
The AD conversion unit 122 includes three AD converters. The AD converter 122 sequentially AD converts each of the three color signals included in the analog color signal AV2 ', and outputs a digital image signal DV2 composed of the three color digital color signals. The digital image signal DV2 and the two synchronization signals VSYNC2 and HSYNC2 are input to the video processor 200. Note that the AD conversion of the analog color signal AV <b> 2 ′ in the AD conversion unit 122 is executed according to a dot clock signal DCLK (described later) output from the video processor 200.
[0033]
The video processor 200 has a function of generating image data to be supplied to the liquid crystal panel 300 from the input digital image signals DV1 and DV2.
[0034]
FIG. 4 is an explanatory diagram showing an example of the internal configuration of the video processor 200. The video processor 200 includes a sampling clock generation unit 210, a data selector 220, a control unit 230, a write control unit 240, a read control unit 250, a read address generation unit 260, and a brightness adjustment unit 280. Yes. The control unit 230 is connected to the CPU 100 via the bus line 100a, and controls each unit in the video processor 200 based on an instruction from the CPU 100. Note that the read control unit 250 and the read address generation unit 260 shown in FIG. 4 correspond to the image distortion adjustment unit of the present invention.
[0035]
The sampling clock generator 210 receives the first synchronization signals VSYNC 1 and HSYNC 1 output from the video decoder 110 of FIG. 3 and the second synchronization signals VSYNC 2 and HSYNC 2 output from the synchronization separation circuit 120. . The sampling clock generation unit 210 selects one of the first and second synchronization signals based on the control signal CTRS supplied from the control unit 230, and selects the sampling clock signal SCLK based on the selected synchronization signal. Generate. The generated sampling clock signal SCLK is input to the control unit 230, and the control unit 230 is used to control each unit in the video processor 200. When the second synchronization signals VSYNC2 and HSYNC2 are selected, the sampling clock generation unit 210 further outputs the dot clock signal DCLK to be supplied to the AD conversion unit 122 (FIG. 3).
[0036]
Two digital image signals DV1 and DV2 are input to the data selector 220. The data selector 220 selects one of the two digital image signals DV1 and DV2 based on the select signal SEL supplied from the control unit 230, and outputs the selected digital image signal DV.
[0037]
The writing control unit 240 has a function of alternately writing the input digital image signal DV into the two sets of frame memories 140 (FIG. 3). The writing control unit 240 writes image data DT and DT ′ for one frame in the two sets of frame memories 140, respectively. That is, after the image data DT for one frame is written in the first frame memory 140, the image data DT ′ for the next frame is written in the second frame memory 142. The image data DT and DT ′ written in the first and second frame memories 140 and 142 represent substantially rectangular images. Note that the write operation in the write control unit 240 is performed based on a write control signal CTRW supplied from the control unit 230. Write control signal CTRW is generated in control unit 230 in accordance with sampling clock signal SCLK input from sampling clock generation unit 210.
[0038]
The read control unit 250 has a function of alternately reading out distorted image data DTS and DTS ′ representing a substantially trapezoidal distorted image from the two sets of frame memories 140 and 142. The readout control unit 250 includes distorted image data DTS representing a substantially trapezoidal distortion image among a plurality of pixels included in the image data DT and DT ′ representing the substantially rectangular image stored in the two sets of frame memories 140 and 142. , DTS ′ is selected and read out from the pixel data necessary for constituting the DTS ′. The selection of the readout pixel is controlled by an address control signal ADD supplied from a readout address generation unit 260 described later. The read operation is controlled by a control signal CTRL supplied from the control unit 230. The control signal CTRL is a signal generated according to the synchronization signals VSYNC3 and HSYNC3 supplied from the drive unit 150 (FIG. 3). The distorted image data DTS and DTS ′ thus read out are sequentially output from the read control unit 250 as distorted image data DDV. If a distortion image based on the distortion image data DDV is formed as it is on the liquid crystal panel 300 of FIG. 3, it is possible to display a normal image 34A without distortion as shown in FIG. It becomes.
[0039]
FIG. 5 is an explanatory diagram showing the relationship between the normal image 34 ′ and the distorted image 34 </ b> A ′ formed on the liquid crystal panel 300. Note that the two images 34 ′ and 34 </ b> A ′ illustrated in FIG. 5 are the same as the two images 34 ′ and 34 </ b> A ′ illustrated in FIG. As described above, when tilting is not performed, the normal image 34 ′ may be formed on the liquid crystal panel 300. On the other hand, when tilting projection is performed, it is necessary to form a distorted image 34 </ b> A ′ on the liquid crystal panel 300. In other words, in the case of tilt projection, it is necessary to convert the normal image 34 ′ into a distorted image 34 </ b> A ′ and form it. Specifically, as shown in FIG. 5, the image is changed so that the image at the point FP ′ of the normal image 34 ′ is formed at the position of the point FPA ′ in the distorted image 34 </ b> A ′. The relationship between the normal image 34 ′ and the distorted image 34 </ b> A ′ is expressed by, for example, Expression 1.
[0040]
[Expression 1]

[0041]
Here, (Xb, Yb) indicates the coordinate value of the normal image 34 ′ in the liquid crystal panel 300 as shown in FIG. 5. (Hn, Vn) indicates the coordinate value of the distorted image 34 </ b> A ′ in the liquid crystal panel 300. Vm represents the maximum coordinate value in the vertical direction of the liquid crystal panel 300. These coordinate values are the coordinate values when the lower left point of the liquid crystal panel 300 shown in FIG. 5 is the coordinate origin ORG (0, 0). L indicates the distance between the liquid crystal panel 300 and the screen 30 in the projection display device 20, as shown in FIG. θ represents a deviation angle between the optical axis 20 c of light emitted from the projection display device 20 and the normal line 30 n of the screen 30. The relationship shown in Formula 1 is described in, for example, Japanese Patent Laid-Open No. 8-98119. In the present embodiment, the distorted image 34A ′ is generated based on the relationship shown in Equation 1, but other relationship equations may be used.
[0042]
The two parameters of the distance L and the deviation angle θ in Equation 1 are given by the parameter measuring unit 402 or the parameter setting unit 404 shown in FIG. The parameter measuring unit 402 has a function of measuring the distance L and the deviation angle θ. The distance L can be measured, for example, by providing the projection display device 20 with an ultrasonic sensor for distance measurement used for autofocusing of a camera. As shown in FIG. 1, the shift angle θ can be obtained by measuring how a “foot” that adjusts the inclination of the projection display device 20 during tilt projection is measured. On the other hand, in the parameter setting unit 404, the user sets values of the distance L and the deviation angle θ. If the two parameters L and θ determined by the parameter measuring unit 402 or the parameter setting unit 404 are used, the positional relationship between the points in the positive image 34 ′ corresponding to the respective points in the distorted image 34A ′ from the relationship of Equation 1. Can know. The two parameters L and θ determined by the parameter measurement unit 402 or the parameter setting unit 404 are given to the read address generation unit 260 via the bus line 100a and the control unit 230 of FIG.
[0043]
The read address generation unit 260 outputs the above-described address control signal ADD based on the relationship of Equation 1 using the values of the parameters L and θ input from the control unit 230. Specifically, an address value for specifying the pixel position (Xb, Yb) of the positive image 34 ′ corresponding to each pixel position (Hn, Vn) included in the substantially trapezoidal distortion image 34A ′ shown in FIG. Is output as an address control signal ADD. Further, the pixel position of the positive image 34 ′ corresponding to the pixel position not included in the substantially trapezoidal distortion image 34 </ b> A ′ indicated by the hatched portion in FIG. 5 is outside the region of the liquid crystal panel 300. In this case, as the address control signal ADD, an address value (for example, “0”) indicating that there is no corresponding pixel of the normal image 34 ′ is output. The address control signal ADD is sequentially output at a timing based on the control signal CTRL supplied from the control unit 230.
[0044]
As described above, the readout control unit 250 is stored at the address specified by the address control signal ADD among the plurality of pixels included in the image data DT and DT ′ stored in the two frame memories 140 and 142. By sequentially extracting only the pixel data, the distorted image data DDV is output. At this time, when an address value (for example, “0”) indicating that the corresponding pixel of the normal image 34 ′ does not exist is input, for example, pixel data indicating “black” is output for the pixel. To do. Accordingly, the read control unit 250 can output distorted image data DDV representing a rectangular image including the substantially trapezoidal distorted image 34A ′ shown in FIG.
[0045]
The luminance adjustment unit 280 has a function of adjusting the luminance of the distorted image represented by the distorted image data DDV output from the read control unit 250 and outputting adjusted image data ADV.
[0046]
FIG. 6 is an explanatory diagram showing processing in the brightness adjustment unit 280. FIG. 6A shows a case where a substantially trapezoidal distortion image represented by the distortion image data DDV input to the luminance adjustment unit 280 is formed on the liquid crystal panel 300 as it is. 6B to 6D show a case where three types of substantially trapezoidal distortion images represented by the adjusted image data ADV1 to ADV3 output from the brightness adjustment unit 280 are formed on the liquid crystal panel 300. FIG. Yes. The images shown in FIGS. 6A to 6D are stored in the frame memories 140 and 142 of FIG. 4 with image data DT and DT ′ representing single-color “solid images” having the same pixel values. 3 shows an image represented by the distorted image data DDV and the adjusted image data ADV1 to ADV3 output in the case where Therefore, the luminance in the image of the distorted image in FIG. 6A formed on the liquid crystal panel 300 is substantially uniform, but the luminance in the images of the adjusted images in FIGS. 6B to 6D is adjusted. ing.
[0047]
The processes shown in FIGS. 6B to 6D in the luminance adjustment unit 280 are performed according to the luminance adjustment mode set in the luminance adjustment mode setting unit 406 in FIG. The brightness adjustment mode set by the user is input to the brightness adjustment unit 280 as the mode information signal MD via the bus line 100a and the control unit 230 as shown in FIG.
[0048]
FIG. 6B shows a case where the adjusted image represented by the adjusted image data ADV1 when the first luminance adjustment mode is used is formed on the liquid crystal panel 300. When the first luminance adjustment mode is used, the horizontal line at the upper base of the substantially trapezoidal distortion image is set as the reference line, and the luminance of each horizontal line is adjusted. Here, the “reference line” means a horizontal line serving as a reference for luminance adjustment, and the luminance of other horizontal lines is adjusted according to the relationship with the reference line. In the luminance adjustment of this embodiment, each of a plurality of pixels included in each horizontal line is multiplied by a first coefficient K1 (t) that changes for each horizontal line. The first coefficient K1 (t) is expressed by the following formula 2.
[0049]
[Expression 2]

[0050]
Here, as shown in FIG. 6A, t indicates a t-th horizontal line when each horizontal line constituting the substantially trapezoidal distortion image is counted from the bottom, and tmax indicates an approximately trapezoidal shape. The tmax-th horizontal line, which is the horizontal line at the top of the distortion image of FIG. H (t) means the width of the t-th horizontal line constituting the substantially trapezoidal distortion image, and V (t) means the height of the t-th horizontal line. Therefore, H (tmax) indicates the width of the tmax-th horizontal line (upper bottom horizontal line) of the substantially trapezoidal distortion image, and V (tmax) indicates the height of the upper bottom horizontal line. Yes. Hm is the same as Hm shown in FIG. 5 and means the width H (0) of the 0th horizontal line (lower bottom horizontal line). S1 is a constant larger than zero. The constant S1 will be further described later.
[0051]
As can be seen from Equation 2, the first coefficient K1 (t) is “S1” for the tmax-th upper-bottom horizontal line, which is the reference line, and “S1 · ( H (tmax) · V (tmax)) / (Hm · Vm) ”. Specifically, the function in the first parenthesis including Hm in Equation 2 is “1” in the tmax-th horizontal line that is the upper bottom line, and “H” in the 0th horizontal line that is the lower base line. It changes linearly according to the horizontal line t so that (tmax) / Hm ". The function in the second parenthesis including Vm is “1” in the tmax-th horizontal line that is the upper bottom line, and “V (tmax) / Vm” in the 0th horizontal line that is the lower-bottom line. So that it changes linearly according to the horizontal line t. Accordingly, the first coefficient K1 (t) decreases monotonously as it goes from the upper bottom horizontal line constituting the substantially trapezoidal distortion image toward the lower bottom horizontal line.
[0052]
As can be seen from the above description, the function in the first parenthesis including Hm in Equation 2 means the luminance correction amount in the horizontal direction based on the horizontal distortion correction amount of the substantially trapezoidal distortion image. . The function in the second parenthesis including Vm in Equation 2 means the luminance correction amount in the vertical direction based on the vertical distortion correction amount of the substantially trapezoidal distortion image.
[0053]
At this time, the luminance of the substantially trapezoidal distortion image formed on the liquid crystal panel 300 is adjusted as shown in FIG. That is, the luminance of the line image of the upper bottom horizontal line that is the reference line displayed on the liquid crystal panel 300 is adjusted to S1 times, and the luminance of the line images of the other horizontal lines is adjusted from the horizontal line on the upper bottom side. As it goes to the horizontal line on the bottom side, it decreases monotonously. In FIG. 6B, for convenience of illustration, the luminance change of the substantially trapezoidal distortion image is divided into six regions, but actually, the luminance changes monotonously for each horizontal line. is doing. In this way, by generating adjusted image data ADV1 in which the luminance of the substantially trapezoidal distortion image formed on the liquid crystal panel 300 is adjusted, the luminance of the positive image displayed on the screen 30 can be set for the entire image. It becomes possible to make it almost uniform.
[0054]
In addition, when the image displayed on the screen 30 is relatively dark, the entire image can be brightened by setting the constant S1 to a value larger than 1. That is, the constant S1 included in the first coefficient K1 (t) is for adjusting the luminance of the entire image while maintaining the luminance of the entire positive image displayed on the screen 30 substantially uniform. It is a multiplier. Note that the setting of the constant S1 can be successfully performed by, for example, obtaining an average pixel value of all the pixels included in the substantially trapezoidal distortion image before adjusting the luminance. This average pixel value can be obtained when each pixel data constituting the substantially trapezoidal distortion image is read by the reading control unit 250 (FIG. 4). When the obtained average pixel value is relatively small, a relatively large value exceeding 1 may be set as the constant S1 in order to brighten the image displayed on the screen 30.
[0055]
FIG. 6C shows a case where the adjusted image represented by the adjusted image data ADV <b> 2 when the second luminance adjustment mode is used is formed on the liquid crystal panel 300. When the second brightness adjustment mode is used, the horizontal line at the bottom of the substantially trapezoidal distortion image is set as the reference line, and the brightness of each horizontal line is adjusted. In this case, each of the plurality of pixels included in each horizontal line is multiplied by a second coefficient K2 (t) that changes for each horizontal line. The second coefficient K2 (t) is expressed by the following mathematical formula 3.
[0056]
[Equation 3]

[0057]
As can be seen from Equation 3, the second coefficient K2 (t) is “S2” for the 0th lower-bottom horizontal line, which is the reference line, and “S2 · (Hm · Vm) / (H (tmax) · V (tmax)) ”. Specifically, the function in the first parenthesis including Hm in Equation 3 is “Hm / H (tmax)” in the tmax-th horizontal line that is the upper bottom line, and the 0th function that is the lower bottom line. It changes linearly according to the horizontal line t so that it becomes “1” in the horizontal line. The function in the second parenthesis including Vm is “Vm / V (tmax)” in the tmax-th horizontal line as the upper bottom line, and “1” in the 0th horizontal line as the lower-bottom line. So that it changes linearly according to the horizontal line t. Accordingly, also in this case, the second coefficient K2 (t) monotonously decreases from the upper bottom horizontal line constituting the substantially trapezoidal distortion image toward the lower bottom horizontal line.
[0058]
At this time, as shown in FIG. 6C, the luminance of the substantially trapezoidal distortion image formed on the liquid crystal panel 300 is adjusted to S2 times for the luminance of the line image of the lower bottom horizontal line as the reference line. The luminances of the line images of other horizontal lines increase monotonously from the lower bottom horizontal line toward the upper bottom horizontal line. Even when the second coefficient K2 (t) is used, the luminance of the positive image displayed on the screen 30 is made substantially uniform over the entire image as in the case where the first coefficient K1 (t) is used. be able to.
[0059]
However, the second coefficient K2 (t) shown in Formula 3 is always 1 or more when the constant S2 is “1”, and is therefore included in the line image on the upper base side of the substantially trapezoidal distortion image. When the pixel value of the pixel to be displayed shows a large value, the dynamic range of the pixel value may be exceeded by multiplying by the second coefficient K2 (t). For example, when the pixel value of each pixel is composed of 8 bits, the value may exceed the dynamic range (0 to 255) by multiplying by the second coefficient K2 (t). In such a case, it is preferable to darken the entire image by setting the constant S2 included in the second coefficient K2 (t) to a value smaller than 1. Note that the setting of the constant S2 can be performed based on the average pixel value of all the pixels included in the substantially trapezoidal distortion image before adjusting the luminance, similarly to the setting of the constant S1 described above.
[0060]
FIG. 6D shows a case where the adjusted image represented by the adjusted image data ADV3 when the third luminance adjustment mode is used is formed on the liquid crystal panel 300. When the third brightness adjustment mode is used, a horizontal line that maximizes the average pixel value of a plurality of pixels included in each horizontal line of the substantially trapezoidal distortion image is set as a reference line, and the brightness of each horizontal line is set. Is adjusted. In the luminance adjustment, each of the plurality of pixels included in each horizontal line is multiplied by a third coefficient K3 (t) that changes for each horizontal line. The third coefficient K3 (t) is expressed by the following mathematical formula 4.
[0061]
[Expression 4]

[0062]
Here, tl indicates a tl-th horizontal line serving as a reference line. H (tl) indicates the width of the tl-th horizontal line constituting the substantially trapezoidal distortion image.
[0063]
As can be seen from Equation 4, the third coefficient K3 (t) decreases monotonically for each horizontal line constituting the substantially trapezoidal distortion image from the upper bottom horizontal line toward the lower bottom horizontal line. In addition, the reference line tl is “S3”.
[0064]
At this time, the luminance of the substantially trapezoidal distortion image formed on the liquid crystal panel 300 is adjusted as shown in FIG. The luminance of the line image of the tl-th horizontal line that is the reference line displayed on the liquid crystal panel 300 is adjusted to S3 times, and the luminance of the line images of the other horizontal lines is adjusted from the horizontal line on the upper bottom side to the lower bottom. It monotonically decreases toward the horizontal line on the side. Therefore, even when the third coefficient K3 (t) is used, the positive image displayed on the screen 30 is the same as when the first and second coefficients K1 (t) and K2 (t) are used. Can be made substantially uniform over the entire image. In this case, the horizontal line that maximizes the average pixel value of a plurality of pixels included in each horizontal line of the substantially trapezoidal distortion image is set as the reference line. Compared to the second luminance adjustment mode (Equation 3), the pixel value included in each horizontal line is less likely to exceed the dynamic range.
[0065]
In the above, the horizontal line that maximizes the average pixel value of a plurality of pixels included in each horizontal line of the substantially trapezoidal distortion image is set as the reference line, but is included in the approximately trapezoidal distortion image. A horizontal line including a pixel having a maximum pixel value among all the pixels may be set as a reference line.
[0066]
Thus, the brightness adjustment unit 280 (FIG. 4) determines the brightness adjustment mode according to the mode information signal MD input from the control unit 230. In addition, it is necessary in each brightness adjustment mode based on coefficient determination information KD such as the width and height of various horizontal lines (upper bottom line, lower bottom line, reference line, etc.) input from the read address generator 260. The contents of the luminance adjustment coefficients K1 (t) to K3 (t) are determined. Based on the determined luminance adjustment coefficient, the pixel value of each line image signal constituting the distorted image data DDV is adjusted, and adjusted image data ADV is output.
[0067]
By the way, in the first to third coefficients K1 (t) to K3 (t), not only the horizontal direction (line direction) but also the vertical direction (from the upper bottom horizontal line to the lower bottom horizontal line) Also in the height direction), the luminance correction amount changes linearly. For example, in the first coefficient K1 (t) shown in Formula 2, the function in the second parenthesis including Vm indicating the luminance correction amount in the vertical direction is a function that changes linearly according to the position t of the horizontal line. It is. In practice, however, the substantially trapezoidal distortion image is corrected as shown in FIG. That is, the substantially trapezoidal distortion image is corrected almost linearly from the upper bottom horizontal line to the lower bottom horizontal line in the horizontal direction, but is corrected in a larger proportion in the vertical direction toward the upper base side. Yes. Therefore, it is desirable to use a correction amount that closely follows Formula 1 for the luminance correction amount in the vertical direction. In this way, the luminance adjustment corresponding to the correction amount when generating the distorted image can be performed, and thus there is an advantage that the luminance distribution in the image displayed on the screen can be made more uniform. However, when the luminance correction amount that linearly changes in the vertical direction as described above is used, there is an advantage that the luminance in the image of the positive image displayed on the screen 30 can be easily made substantially uniform. .
[0068]
The coefficients K1 (t) to K3 (t) may be determined based only on the luminance correction amount in the horizontal direction of each line constituting the substantially trapezoidal distortion image. In this case, for the first coefficient K1 (t), for example, a function obtained by squaring the function in the first parenthesis including Hm related to the luminance correction amount in the horizontal direction of Equation 2 may be used. Even in this case, the luminance distribution of the positive image displayed on the screen 30 can be made substantially uniform.
[0069]
Further, instead of the first to third coefficients K1 (t) to K3 (t), a coefficient K (t) represented by the following formula 5 may be used.
[0070]
[Equation 5]

[0071]
Here, s is a constant larger than 0.
[0072]
Even using the above Equation 5, there is a possibility that the luminance distribution of the positive image displayed on the screen 30 can be made substantially uniform.
[0073]
As can be seen from the above description, the luminance adjustment coefficient only needs to be determined based on the distortion correction amount used when generating the distortion image data DDV representing the substantially rectangular distortion image.
[0074]
The driving unit 150 in FIG. 3 drives the liquid crystal panel 300 according to the supplied adjusted image data ADV. The liquid crystal panel 300 is irradiated with light emitted from the light source unit 310, and the liquid crystal panel 300 is driven to generate a distorted image in which the luminance represented by the adjusted image data ADV is adjusted. It is formed. The distorted image formed on the liquid crystal panel 300 is projected by the projection optical system, and a positive image whose luminance is adjusted almost uniformly is displayed on the screen 30.
[0075]
As described above, in this embodiment, the brightness of the line image of each line constituting the substantially trapezoidal distortion image formed on the liquid crystal panel 300 is adjusted. As a result, a positive image without distortion can be displayed on the screen 30, and the luminance of the displayed positive image can be well adjusted.
[0076]
The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.
[0077]
(1) In the above-described embodiment, the liquid crystal panel 300 is used as the image forming unit. However, instead of this, a micromirror type light modulation device may be used. For example, a DMD (digital micromirror device) (registered trademark of TI) can be used as the micromirror light modulator. Further, a high-intensity CRT can be used as the image forming unit. As the image forming unit of the present invention, various devices that convert image signals into image light can be generally used.
[0078]
Note that when the above-described high-intensity CRT is used as the image forming unit, the maximum luminance state is not normally used. Therefore, the luminance of the line image on the upper base side of the substantially trapezoidal distortion image is set to the corresponding rectangular shape. It is easy to make it larger than the brightness of the line image of the positive image. On the other hand, in a liquid crystal panel, since the panel is usually used up to the highest light transmittance, the brightness of the line image on the upper base side of the substantially trapezoidal distortion image is set to the line image of the corresponding rectangular positive image. It is preferable that the luminance of the line image on the lower base side of the substantially trapezoidal distortion image is reduced to be equal to the luminance.
[0079]
(2) In the above embodiment, the readout control unit 250 shown in FIG. 4 outputs the distorted image data DDV by reading out the pixel data necessary for constructing the substantially trapezoidal distorted image from the frame memories 140 and 142 as they are. However, each pixel data included in the substantially trapezoidal distortion image data DDV may be complemented. Complementation of pixel data may be performed between pixel data of adjacent pixels in a substantially trapezoidal distortion image, for example.
[0080]
(3) In the above embodiment, as shown in FIG. 1B, the case where the image emitted from the projection display device 20 is projected upward is described. However, the projection display device 20 is suspended on a ceiling or the like. The present invention can be similarly applied to a case where the image is lowered and projected downward. In this case, an image that is upside down with the substantially trapezoidal distortion image shown in FIG. In this case as well, the luminance of the substantially trapezoidal distortion image may be adjusted so as to monotonously decrease from the upper bottom (short side) horizontal line toward the lower base (long side) horizontal line.
[0081]
(4) In the above embodiment, as shown in FIG. 4, the read control unit 250, the read address generation unit 260, and the luminance adjustment unit 280 corresponding to the image distortion adjustment unit and the luminance adjustment unit of the present invention are the projection type. Although the case where it is provided inside the display device 20 has been described, another image processing device may include an image distortion adjustment unit and a luminance adjustment unit. In this case, another image processing device may generate an adjusted image signal by adjusting the input image signal and supply the adjusted image signal to a projection unit including an image forming unit and a projection optical system. In this case, the other image processing device and the projection unit correspond to the projection display device of the present invention.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram showing an image projected using a projection display device.
FIG. 2 is an explanatory diagram showing a relationship between an image displayed on the screen 30 and an image formed on the liquid crystal panel 300 when tilting projection is performed.
FIG. 3 is a block diagram mainly showing an electrical configuration of a projection display apparatus as an embodiment of the present invention.
4 is an explanatory diagram showing an example of an internal configuration of a video processor 200. FIG.
FIG. 5 is an explanatory diagram showing a relationship between a normal image 34 ′ and a distorted image 34 A ′ formed on the liquid crystal panel 300.
FIG. 6 is an explanatory diagram showing processing in a luminance adjustment unit 280.
[Explanation of symbols]
20. Projection display device
30 ... Screen
100 ... CPU
100a ... Bus line
110: Video decoder
120 ... Sync separation circuit
122 ... AD converter
140, 142 ... frame memory
150 ... Drive unit
200: Video processor
210: Sampling clock generator
220: Data selector
230 ... Control unit
240: Write control unit
250 ... Reading control unit
260 ... Read address generator
280: Brightness adjustment unit
300 ... LCD panel
310 ... Light source section
320 ... Projection lens
402: Parameter measurement unit
404 ... Parameter setting section
406 ... Brightness adjustment mode setting unit

Claims (9)

A projection display device,
An image adjustment unit for adjusting an input image signal and outputting an adjusted image signal;
An image forming unit for forming an image based on the adjusted image signal;
A projection optical system for projecting an original image formed on the image forming unit onto a screen;
With
The image adjustment unit
In order to correct image distortion that occurs when a substantially rectangular original image is projected on the screen, the input image signal is adjusted so that a substantially trapezoidal original image is formed in the image forming unit. An image distortion adjustment unit;
A plurality of line images constituting the substantially trapezoidal original image in order to correct a luminance change in the image generated when the substantially trapezoidal original image formed in the image forming unit is projected on the screen. luminance, such that monotonically decreases from the line image of the upper base of the substantially trapezoidal original image toward the line image of the lower bottom, the pixel values of a plurality of pixels included in, respectively that of the plurality of line images A brightness adjustment unit for adjusting
With
Said adjustment before Symbol pixel value by the luminance adjusting unit, a projection display, characterized in that it is performed based on the correction amount of distortion that is used to adjust the image signal to be the input in the image distortion adjustment unit apparatus. The projection display device according to claim 1,
The luminance adjustment unit uses the upper-bottom line image among the plurality of line images included in the substantially trapezoidal original image as a reference line image, and each of the luminance adjustment coefficients decreases monotonously as it goes to the lower-bottom line image. Multiply the pixel values of multiple pixels in the line image to adjust the brightness,
The projection display device, wherein the luminance adjustment coefficient is determined based on a correction amount of the distortion. The projection display device according to claim 1,
The luminance adjustment unit uses a lower-bottom line image among the plurality of line images included in the substantially trapezoidal original image as a reference line image, and sets a luminance adjustment coefficient that increases monotonously toward the upper-bottom line image. Multiply the pixel values of multiple pixels in the line image to adjust the brightness,
The projection display device, wherein the luminance adjustment coefficient is determined based on a correction amount of the distortion. The projection display device according to claim 1,
The brightness adjustment unit uses, as a reference line image, a line image in which an average of pixel values of a plurality of pixels included in each line image among the plurality of line images included in the substantially trapezoidal original image is the maximum. The brightness is adjusted by multiplying the pixel value of a plurality of pixels included in each line image by a brightness adjustment coefficient that increases monotonously as it goes to the upper bottom line image from the reference line image, and the lower bottom line image from the reference line is adjusted. Multiply the pixel values of multiple pixels in each line image by a brightness adjustment factor that decreases monotonously as you go, and adjust the brightness.
The projection display device, wherein the luminance adjustment coefficient is determined based on a correction amount of the distortion. A projection display device according to any one of claims 2 to 4 ,
The projection display device, wherein the luminance adjustment coefficient is determined using a correction amount of distortion in a height direction of the substantially trapezoidal original image. The projection display device according to claim 5 ,
The projection display device, wherein the brightness adjustment coefficient is further determined using a correction amount of distortion in a line direction of the substantially trapezoidal original image. The projection display device according to claim 6 ,
The brightness adjustment coefficient is
From the upper-bottom line image determined based on the width of the upper-bottom line image of the substantially trapezoidal original image formed in the image forming unit and the width of the lower-bottom line image of the substantially rectangular original image It is determined based on the first adjustment coefficient that changes linearly toward the bottom-bottom line image, the height of the substantially trapezoidal original image formed in the image forming unit, and the height of the substantially rectangular original image. A projection display device, which is a value proportional to a value obtained by multiplying a second adjustment coefficient that linearly changes from the upper-bottom line image toward the lower-bottom line image. A projection display device according to any one of claims 2 to 7 ,
The projection display device, wherein the brightness adjustment coefficient includes a multiplier for adjusting brightness in an image displayed on the screen over the entire image. An image processing method for processing an original image formed in an image forming unit of a projection display device that projects and displays an image on a screen,
(A) In order to correct image distortion that occurs when a substantially rectangular original image is projected on the screen, an image signal input so that a substantially trapezoidal original image is formed in the image forming unit. Adjusting, and
(B) A plurality of the substantially trapezoidal original images are formed in order to correct a luminance change in the image that occurs when the substantially trapezoidal original image formed in the image forming unit is projected on the screen. a plurality of pixel intensity of the line image, the generally such that monotonically decreases toward the line image of the lower bottom trapezoidal in the original image the upper base of the line image, contained in, respectively that of the plurality of line images of Adjusting the pixel value of
With
Said adjustment before Symbol pixel values in step (b), image processing, characterized in that it is performed based on the correction amount of distortion that is used to adjust the image signal to be the input in the step (a) Method.

Images