JP4649786B2 - Coefficient data generating apparatus and generating method, information signal processing apparatus and processing method using the same, and coefficient seed data generating apparatus and generating method used therefor - Google Patents

Description

【００４５】
そして、動きクラス検出回路１２５では、上述したように算出された平均値ＡＶが１個または複数個のしきい値と比較されて動きクラスのクラス情報ＭＶが得られる。例えば、３個のしきい値ｔｈ１，ｔｈ２，ｔｈ３（ｔｈ１＜ｔｈ２＜ｔｈ３）が用意され、４つの動きクラスを検出する場合、ＡＶ≦ｔｈ１のときはＭＶ＝０、ｔｈ１＜ＡＶ≦ｔｈ２のときはＭＶ＝１、ｔｈ２＜ＡＶ≦ｔｈ３のときはＭＶ＝２、ｔｈ３＜ＡＶのときはＭＶ＝３とされる。
【００４６】
また、画像信号処理部１１０は、空間クラス検出回路１２４より出力される空間クラスのクラス情報としての再量子化コードｑｉと、動きクラス検出回路１２５より出力される動きクラスのクラス情報ＭＶに基づき、作成すべきＨＤ信号（１０５０ｉ信号）の画素（注目画素）が属するクラスを示すクラスコードＣＬを得るためのクラス合成回路１２６を有している。
このクラス合成回路１２６では、（３）式によって、クラスコードＣＬの演算が行われる。なお、（３）式において、Ｎａは空間クラスタップのデータ（ＳＤ画素データ）の個数、ＰはＡＤＲＣにおける再量子化ビット数を示している。
【００４７】
【数２】

【００４８】
また、画像信号処理部１１０は、係数メモリ１３４を有している。この係数メモリ１３４は、後述する推定予測演算回路１２７で使用される推定式の複数の係数データＷｉを、クラス毎に、格納するものである。この係数データは、ＳＤ信号（５２５ｉ信号）を、ＨＤ信号（１０５０ｉ信号）に変換するための情報である。係数メモリ１３４には上述したクラス合成回路１２６より出力されるクラスコードＣＬが読み出しアドレス情報として供給され、この係数メモリ１３４からはクラスコードＣＬに対応した推定式の係数データＷｉ（ｉ＝１〜ｎ）が読み出され、推定予測演算回路１２７に供給されることとなる。
【００４９】
また、画像信号処理部１１０は、情報メモリバンク１３５を有している。後述する推定予測演算回路１２７では、予測タップのデータ（ＳＤ画素データ）ｘｉと、係数メモリ１３４より読み出される係数データＷｉとから、（４）式の推定式によって、作成すべきＨＤ画素データｙが演算される。第１のタップ選択回路１２１で選択される予測タップ数が１０個であるとき、（４）式におけるｎは１０となる。
【００５０】
【数３】

【００５１】
情報メモリバンク１３５には、この推定式の係数データＷｉ（ｉ＝１〜ｎ）を生成する際に用いられる第１の係数種データとしての係数種データｗ1_i0〜ｗ1_i9（ｉ＝１〜ｎ）と、第２の係数種データとしての係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9（ｉ＝１〜ｎ）が、記憶されている。
【００５２】
上述したように、５２５ｉ信号を１０５０ｉ信号に変換する場合、奇数、偶数のそれぞれのフィールドにおいて、５２５ｉ信号の１画素に対応して１０５０ｉ信号の４画素を得る必要がある。この場合、奇数、偶数のそれぞれのフィールドにおける１０５０ｉ信号を構成する２×２の単位画素ブロック内の４画素は、それぞれ中心予測タップに対して異なる位相ずれを持っている。
【００５３】
図２は、奇数フィールドにおける１０５０ｉ信号を構成する２×２の単位画素ブロック内の４画素ＨＤ₁〜ＨＤ₄における中心予測タップＳＤ₀からの位相ずれを示している。ここで、ＨＤ₁〜ＨＤ₄の位置は、それぞれ、ＳＤ₀の位置から水平方向にｋ₁〜ｋ₄、垂直方向にｍ₁〜ｍ₄だけずれている。
【００５４】
図３は、偶数フィールドにおける１０５０ｉ信号を構成する２×２の単位画素ブロック内の４画素ＨＤ₁′〜ＨＤ₄′における中心予測タップＳＤ₀′からの位相ずれを示している。ここで、ＨＤ₁′〜ＨＤ₄′の位置は、それぞれ、ＳＤ₀′の位置から水平方向にｋ₁′〜ｋ₄′、垂直方向にｍ₁′〜ｍ₄′だけずれている。
【００５５】
したがって、上述した情報メモリバンク１３５には、クラスおよび出力画素（ＨＤ₁〜ＨＤ₄，ＨＤ₁′〜ＨＤ₄′）の組み合わせ毎に、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9（ｉ＝１〜ｎ）が格納されている。
【００５６】
図４は、係数種データｗ1_i0〜ｗ1_i9の生成方法を示している。
まず、ＨＤ信号から複数のＳＤ信号を生成する。この場合、ＨＤ信号からＳＤ信号を得る際に使用する落としフィルタの水平周波数特性と垂直周波数特性を可変するパラメータｈ，ｖを、それぞれ４，５，６，７，８に５段階に可変して、合計２５種類のＳＤ信号を生成する。次に、生成された複数のＳＤ信号とＨＤ信号との間で学習を行って係数種データｗ1_i0〜ｗ1_i9を生成する。
【００５７】
図５は、パラメータｈ，ｖを４，５，６，７，８としたときの落としフィルタにおける水平、垂直の周波数特性を示している。このとき、パラメータｈ，ｖの値が「１」だけ変化する毎に、第１の物理的な特徴量としての帯域が一定幅だけ変化し、パラメータｈ，ｖの値が小さくなるほど帯域は狭くなる。なお、画像のサンプリング周波数をｆsとし、フィルタのサンプリング周波数をｆとするとき、横軸は２ｆ／ｆsを表している。因に、パラメータｈ，ｖが６のときの落としフィルタは、ゲイン０．５において２ｆ／ｆsが０．５であるため、ハーフバンドフィルタを構成している。
【００５８】
図５に示す周波数特性の落としフィルタはいずれもある周波数ｆまで通過させる構成であるため、これらのフィルタを使うと、周波数ｆより小さな周波数帯の信号成分はすべて通過し、周波数ｆより大きい周波数帯の信号成分は通過できないことになる。したがって、これらのフィルタによって、画像のエッジ部分のボケ具合が異なる生徒信号が生成され、これらの生徒信号を使って学習をすると、エッジを強調するような推定式の係数データＷｉを得るための係数種データｗ1_i0〜ｗ1_i9を得ることができる。
【００５９】
係数種データｗ1_i0〜ｗ1_i9の求め方を説明する。
ここでは、上述した（４）式の推定式の係数データＷｉ（ｉ＝１〜ｎ）を、（５）式の近似式で表現する。この（５）式における係数データｗ_i0〜ｗ_i9を求める。

【００６０】
ここで、以下の説明のため、（６）式のように、ｔｊ（ｊ＝０〜９）を定義する。

この（６）式を用いると、（５）式は、（７）式のように書き換えられる。
【００６１】
【数４】

【００６２】
最終的に、上述したように生成された２５種類のＳＤ信号（図４参照）を使用した学習によって、未定係数ｗ_ijを求める。すなわち、クラスおよび出力画素毎に、複数のＳＤ画素データとＨＤ画素データを用いて、二乗誤差を最小にする係数値を決定する。いわゆる最小二乗法による解法である。学習数をｍ、ｋ（１≦ｋ≦ｍ）番目の学習データにおける残差をｅ_k、二乗誤差の総和をＥとすると、（４）式および（５）式を用いて、Ｅは（８）式で表される。ここで、ｘ_ikはＳＤ画像のｉ番目の予測タップ位置におけるｋ番目の画素データ、ｙ_kはそれに対応するｋ番目のＨＤ画像の画素データを表している。
【００６３】
【数５】

【００６４】
最小二乗法による解法では、（８）式のｗ_ijによる偏微分が０になるようなｗ_ijを求める。これは、（９）式で示される。
【００６５】
【数６】

【００６６】
以下、（１０）式、（１１）式のように、Ｘ_ipjq、Ｙ_ipを定義すると、（９）式は、行列を用いて（１２）式のように書き換えられる。
【００６７】
【数７】

【００６８】
【数８】

【００６９】
この方程式は一般に正規方程式と呼ばれている。この正規方程式は、掃き出し法（Gauss-Jordanの消去法）等を用いて、ｗ_ijについて解かれ、係数データｗ_i0〜ｗ_i9が算出される。このように算出された係数データｗ_i0〜ｗ_i9が係数種データｗ1_i0〜ｗ1_i9となる。
このように求められる係数種データｗ1_i0〜ｗ1_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１２に示すような、基底曲面を表すものとなる。
【００７０】
図６は、係数種データｗ2a_i0〜ｗ2a_i9の生成方法を示している。
まず、ＨＤ信号から複数のＳＤ信号を生成する。この場合、ＨＤ信号からＳＤ信号を得る際に使用する落としフィルタの水平周波数特性を可変するパラメータｈを４，５，６，７，８に５段階に可変し、この落としフィルタの垂直周波数特性を可変するパラメータｖを１，２，３，４，５，６の６段階に可変して、合計３０種類のＳＤ信号を生成する。次に、生成された複数のＳＤ信号とＨＤ信号との間で学習を行って係数種データｗ2a_i0〜ｗ2a_i9を生成する。
【００７１】
図７は、パラメータｖを１，２，３，４，５，６としたときの落としフィルタにおける垂直周波数特性を示している。このとき、パラメータｖの値が「１」だけ変化する毎に、第２の物理的な特徴量としての肩特性が一定幅だけ変化し、パラメータｖの値が小さくなるほど肩部分が滑らかになる。なお、画像のサンプリング周波数をｆsとし、フィルタのサンプリング周波数をｆとするとき、横軸は２ｆ／ｆsを表している。
【００７２】
図７に示す周波数特性の落としフィルタはいずれも高域をある程度通過させながら、低域の通過部分を狭める構成であるため、これらのフィルタを使うと、エッジ部分が残りつつ、画像全体がうっすらとボケたようになる。したがって、これらのフィルタによって、画像全体のボケ具合が異なる生徒信号が生成され、これらの生徒信号を使って学習をすると、画像全体を強調するような推定式の係数データＷｉを得るための係数種データｗ2a_i0〜ｗ2a_i9を得ることができる。
なお、パラメータｈを４，５，６，７，８としたときの落としフィルタにおける水平周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合と同様である（図５、図７の破線参照）。
【００７３】
ここで、ｖ＝６としたときの落としフィルタにおける垂直周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合においてｖ＝６としたときの落としフィルタにおける垂直周波数特性と等しいものとされる。これにより、帯域（第１の物理的な特徴量）が所定値（ｖ＝６）であるときの周波数特性より、肩特性（第２の物理的な特徴量）を変化させて行くことになる。
【００７４】
係数種データｗ2a_i0〜ｗ2a_i9の求め方は、詳細説明は省略するが、上述した係数種データｗ1_i0〜ｗ1_i9を求める場合と同様である。ただし、ここでは、上述したように生成された３０種類のＳＤ信号（図６参照）を使用した学習によって、未定係数ｗ_ijを求めることとなる。
このように求められる係数種データｗ2a_i0〜ｗ2a_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１２に示すような、派生曲面３を表すものとなる。
【００７５】
図８は、係数種データｗ2b_i0〜ｗ2b_i9の生成方法を示している。
まず、ＨＤ信号から複数のＳＤ信号を生成する。この場合、ＨＤ信号からＳＤ信号を得る際に使用する落としフィルタの水平周波数特性を可変するパラメータｈを４，５，６，７，８に５段階に可変し、この落としフィルタの垂直周波数特性を可変するパラメータｖを０，１，２，３，４，５の６段階に可変して、合計３０種類のＳＤ信号を生成する。次に、生成された複数のＳＤ信号とＨＤ信号との間で学習を行って係数種データｗ2b_i0〜ｗ2b_i9を生成する。
【００７６】
図９は、パラメータｖを０，１，２，３，４，５としたときの落としフィルタにおける垂直周波数特性を示している。このとき、パラメータｖの値が「１」だけ変化する毎に、第２の物理的な特徴量としての肩特性が一定幅だけ変化し、パラメータｖの値が小さくなるほど肩部分が滑らかになる。なお、画像のサンプリング周波数をｆsとし、フィルタのサンプリング周波数をｆとするとき、横軸は２ｆ／ｆsを表している。なお、パラメータｈを４，５，６，７，８としたときの落としフィルタにおける水平周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合と同様である（図５、図９の破線参照）。
【００７７】
ここで、ｖ＝５としたときの落としフィルタにおける垂直周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合においてｖ＝５としたときの落としフィルタにおける垂直周波数特性と等しいものとされる。これにより、帯域（第１の物理的な特徴量）が所定値（ｖ＝５）であるときの周波数特性より、肩特性（第２の物理的な特徴量）を変化させて行くことになる。
【００７８】
係数種データｗ2b_i0〜ｗ2b_i9の求め方は、詳細説明は省略するが、上述した係数種データｗ1_i0〜ｗ1_i9を求める場合と同様である。ただし、ここでは、上述したように生成された３０種類のＳＤ信号（図８参照）を使用した学習によって、未定係数ｗ_ijを求めることとなる。
このように求められる係数種データｗ2b_i0〜ｗ2b_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１２に示すような、派生曲面２を表すものとなる。
【００７９】
図１０は、係数種データｗ2c_i0〜ｗ2c_i9の生成方法を示している。
まず、ＨＤ信号から複数のＳＤ信号を生成する。この場合、ＨＤ信号からＳＤ信号を得る際に使用する落としフィルタの水平周波数特性を可変するパラメータｈを４，５，６，７，８に５段階に可変し、この落としフィルタの垂直周波数特性を可変するパラメータｖを−１，０，１，２，３，４の６段階に可変して、合計３０種類のＳＤ信号を生成する。次に、生成された複数のＳＤ信号とＨＤ信号との間で学習を行って係数種データｗ2c_i0〜ｗ2c_i9を生成する。
【００８０】
図１１は、パラメータｖを−１，０，１，２，３，４としたときの落としフィルタにおける垂直周波数特性を示している。このとき、パラメータｖの値が「１」だけ変化する毎に、第２の物理的な特徴量としての肩特性が一定幅だけ変化し、パラメータｖの値が小さくなるほど肩部分が滑らかになる。なお、画像のサンプリング周波数をｆsとし、フィルタのサンプリング周波数をｆとするとき、横軸は２ｆ／ｆsを表している。なお、パラメータｈを４，５，６，７，８としたときの落としフィルタにおける水平周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合と同様である（図５、図１１の破線参照）。
【００８１】
ここで、ｖ＝４としたときの落としフィルタにおける垂直周波数特性は、上述した係数種データｗ1_i0〜ｗ1_i9を生成する場合においてｖ＝４としたときの落としフィルタにおける垂直周波数特性と等しいものとされる。これにより、帯域（第１の物理的な特徴量）が所定値（ｖ＝４）であるときの周波数特性より、肩特性（第２の物理的な特徴量）を変化させて行くことになる。
【００８２】
係数種データｗ2c_i0〜ｗ2c_i9の求め方は、詳細説明は省略するが、上述した係数種データｗ1_i0〜ｗ1_i9を求める場合と同様である。ただし、ここでは、上述したように生成された３０種類のＳＤ信号（図１０参照）を使用した学習によって、未定係数ｗ_ijを求めることとなる。
このように求められる係数種データｗ2c_i0〜ｗ2c_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１２に示すような、派生曲面１を表すものとなる。
【００８３】
上述した係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9は、いずれも肩特性（第２の物理的な特徴量）によって垂直周波数特性を変化させたフィルタで生成された生徒信号から求めたものであり、派生曲面１〜３の曲面形状は似通ったものとなる。
【００８４】
なお、パラメータｈを固定することで、図１２に示すような係数空間は、図１３に示すように平面で表すことができる。図１４は、落としフィルタの帯域、肩特性と係数データＷｉとの対応関係を示している。左側の係数データＡ〜Ｉが、それぞれ右側のＡ〜Ｉの帯域、肩特性に対応している。
【００８５】
また、図１に戻って、画像信号処理部１１０は、情報メモリバンク１３５に格納されている係数種データに基づいて、クラスおよび出力画素の組み合わせ毎に、調整された値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを演算で求める係数生成回路１３６とを有している。
【００８６】
ここで、係数生成回路１３６における推定式の係数データＷｉの演算について説明する。
図１５は、係数空間における係数データＷｉの指定の流れを示している。まず、値ｈ₁，ｖ₁によって基底曲面上の点を指定し、次に、値ｈ₂，ｖ₂によってこの指定された基底曲面上の点を含む派生曲面上の点を指定する。この指定された派生曲面上の点が、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した、求めるべき係数データＷｉとなる。なお、上述したように、肩特性を変化させる方向を垂直方向として、係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9を求めているので、ｈ₁＝ｈ₂である。
【００８７】
値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した係数データＷｉの求め方を説明する。
まず、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応した第２の係数種データが情報メモリバンク１３５に存在する場合について説明する。この場合、当該第２の係数種データを、（５）式の係数データｗ_i0〜ｗ_i9に適用すると共に、この（５）式のパラメータｈ，ｖにｈ₂，ｖ₂を代入することで、係数データＷｉを求める。
【００８８】
次に、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応した第２の係数種データが情報メモリバンク１３５に存在していない場合について説明する。この場合、補間処理によって、係数データＷｉを求める。
【００８９】
図１６に示すように、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）が派生曲面２，３の間に位置する場合を例にとり、図１７を参照して補間処理を説明する。
この場合、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した係数データＷｉをＸとすると、Ｘは（１３）式で算出される。
Ｘ＝（Ｗａ＊ｂ＋Ｗｂ＊ａ）／（ａ＋ｂ） ・・・（１３）
【００９０】
ここで、Ｗａは、派生曲面３に対応した係数種データｗ2a_i0〜ｗ2a_i9を（５）式の係数データｗ_i0〜ｗ_i9に適用すると共に、この（５）式のパラメータｈ，ｖにｈ₂，ｖaを代入することで、求められる係数データＷｉである。ただし、ｖa＝ｖ₂＋（６−ｖ₁）である。
【００９１】
また、Ｗｂは、派生曲面２に対応した係数種データｗ2b_i0〜ｗ2b_i9を（５）式の係数データｗ_i0〜ｗ_i9に適用すると共に、この（５）式のパラメータｈ，ｖにｈ₂，ｖbを代入することで、求められる係数データＷｉである。ただし、ｖb＝ｖ₂−（ｖ₁−５）である。
【００９２】
また、基底曲面に対応した係数種データｗ1_i0〜ｗ1_i9を（５）式の係数データｗ_i0〜ｗ_i9に適用し、この（５）式のパラメータｈ，ｖにそれぞれ「ｈ₁，５」、「ｈ₁，ｖ₁」、「ｈ₁，６」を代入して得られる係数データＷｉをＷ₅，Ｗ_v1，Ｗ₆とするとき、ａ＝Ｗ₆−Ｗ_v1、ｂ＝Ｗ_v1−Ｗ₅である。
【００９３】
なお、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）が派生曲面１，２の間に位置する場合も、同様の補間処理によって、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した係数データＷｉを求めることができる。ただし、着目派生曲面が派生曲面１や派生曲面３の外側に位置する場合には、この補間処理を使用できない。したがって、本実施の形態においては、例えば着目派生曲面が派生曲面１〜派生曲面３の間に位置するように値ｖ₁の変化範囲が制限されるが、着目派生曲面が派生曲面１や派生曲面３の外側に位置する場合には、後述する平行移動処理を適用するようにしてもよい。
【００９４】
図１に戻って、係数生成回路１３６による推定式の係数データＷｉの生成は、例えば各垂直ブランキング期間で行われる。これにより、ユーザのリモコン送信機２００の操作によって値ｈ₁，ｖ₁，ｈ₂，ｖ₂が変更されても、係数メモリ１３４に格納される各クラスの係数データＷｉを、その値に対応したものに即座に変更でき、ユーザによる画質調整がスムーズに行われる。
【００９５】
また、画像信号処理部１１０は、係数生成回路１３６で求められる係数データＷｉ（ｉ＝１〜ｎ）に対応した正規化係数Ｓを、（１４）式によって、演算する正規化係数演算部１３７と、この正規化係数Ｓを格納する正規化係数メモリ１３８とを有している。正規化係数メモリ１３８には上述したクラス合成回路１２６より出力されるクラスコードＣＬが読み出しアドレス情報として供給され、この正規化係数メモリ１３８からはクラスコードＣＬに対応した正規化係数Ｓが読み出され、後述する正規化演算回路１２８に供給されることとなる。
【００９６】
【数９】

【００９７】
また、画像信号処理部１１０は、第１のタップ選択回路１２１で選択的に取り出される予測タップのデータ（ＳＤ画素データ）ｘｉと、係数メモリ１３４より読み出される係数データＷｉとから、（４）式の推定式によって、作成すべきＨＤ信号の画素（注目画素）のデータを演算する推定予測演算回路１２７を有している。
【００９８】
上述したように、ＳＤ信号（５２５ｉ信号）をＨＤ信号（１０５０ｉ信号）に変換する際には、ＳＤ信号の１画素に対してＨＤ信号の４画素（図２のＨＤ₁〜ＨＤ₄、図３のＨＤ₁′〜ＨＤ₄′参照）を得る必要があることから、この推定予測演算回路１２７では、ＨＤ信号を構成する２×２の単位画素ブロック毎に、画素データが生成される。すなわち、この推定予測演算回路１２７には、第１のタップ選択回路１２１より単位画素ブロック内の４画素（注目画素）に対応した予測タップのデータｘｉと、係数メモリ１３４よりその単位画素ブロックを構成する４画素に対応した係数データＷｉとが供給され、単位画素ブロックを構成する４画素のデータｙ₁〜ｙ₄は、それぞれ個別に上述した（４）式の推定式で演算される。
【００９９】
また、画像信号処理部１１０は、推定予測演算回路１２７より順次出力される４画素のデータｙ₁〜ｙ₄を、正規化係数メモリ１３８より読み出される、それぞれの演算に使用された係数データＷｉ（ｉ＝１〜ｎ）に対応した正規化係数Ｓで除算して正規化する正規化演算回路１２８を有している。上述せずも、係数生成回路１３６で求められる推定式の係数データＷｉを求めるものであるが、求められる係数データは丸め誤差を含み、係数データＷｉ（ｉ＝１〜ｎ）の総和が１．０になることは保証されない。そのため、推定予測演算回路１２７で演算される各画素のデータｙ₁〜ｙ₄は、丸め誤差によってレベル変動したものとなる。上述したように、正規化演算回路１２８で正規化することで、その変動を除去できる。
【０１００】
また、画像信号処理部１１０は、正規化演算回路１２８で正規化されて順次供給される単位画素ブロック内の４画素のデータｙ₁′〜ｙ₄′を線順次化して１０５０ｉ信号のフォーマットで出力する後処理回路１２９を有している。
【０１０１】
次に、画像信号処理部１１０の動作を説明する。
バッファメモリ１０９に記憶されているＳＤ信号（５２５ｉ信号）より、第２のタップ選択回路１２２で、作成すべきＨＤ信号（１０５０ｉ信号）を構成する単位画素ブロック内の４画素（注目画素）の周辺に位置する空間クラスタップのデータ（ＳＤ画素データ）が選択的に取り出される。この第２のタップ選択回路１２２で選択的に取り出される空間クラスタップのデータ（ＳＤ画素データ）は空間クラス検出回路１２４に供給される。この空間クラス検出回路１２４では、空間クラスタップのデータとしての各ＳＤ画素データに対してＡＤＲＣ処理が施されて空間クラス（主に空間内の波形表現のためのクラス分類）のクラス情報としての再量子化コードｑｉが得られる（（１）式参照）。
【０１０２】
また、バッファメモリ１０９に記憶されているＳＤ信号（５２５ｉ信号）より、第３のタップ選択回路１２３で、作成すべきＨＤ信号（１０５０ｉ信号）を構成する単位画素ブロック内の４画素（注目画素）の周辺に位置する動きクラスタップのデータ（ＳＤ画素データ）が選択的に取り出される。この第３のタップ選択回路１２３で選択的に取り出される動きクラスタップのデータ（ＳＤ画素データ）は動きクラス検出回路１２５に供給される。この動きクラス検出回路１２５では、動きクラスタップのデータとしての各ＳＤ画素データより動きクラス（主に動きの程度を表すためのクラス分類）のクラス情報ＭＶが得られる。
【０１０３】
この動き情報ＭＶと上述した再量子化コードｑｉはクラス合成回路１２６に供給される。このクラス合成回路１２６では、これら動き情報ＭＶと再量子化コードｑｉとから、作成すべきＨＤ信号（１０５０ｉ信号）を構成する単位画素ブロック毎にその単位画素ブロック内の４画素（注目画素）が属するクラスを示すクラスコードＣＬが得られる（（３）式参照）。そして、このクラスコードＣＬは、係数メモリ１３４および正規化係数メモリ１３８に読み出しアドレス情報として供給される。
【０１０４】
係数メモリ１３４には、例えば各垂直ブランキング期間に、係数生成回路１３６で、ユーザによって調整された値ｈ₁，ｖ₁，ｈ₂，ｖ₂の値に対応して、クラスおよび出力画素（ＨＤ₁〜ＨＤ₄，ＨＤ₁′〜ＨＤ₄′）の組み合わせ毎に、推定式の係数データＷｉ（ｉ＝１〜ｎ）が求められて格納される。この場合、係数生成回路１３６では、情報メモリバンク１３５に格納されている第１の係数種データとしての係数種データｗ1_i0〜ｗ1_i9（ｉ＝１〜ｎ）と、第２の係数種データとしての係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9（ｉ＝１〜ｎ）とに基づいて、上述したように補間処理等によって、推定式の係数データＷｉが求められる。
また、正規化係数メモリ１３８には、上述したように係数生成回路１３６で求められた推定式の係数データＷｉ（ｉ＝１〜ｎ）に対応した正規化係数Ｓが正規化係数演算部１３７で生成されて格納される。
【０１０５】
係数メモリ１３４に上述したようにクラスコードＣＬが読み出しアドレス情報として供給されることで、この係数メモリ１３４からクラスコードＣＬに対応した４出力画素（奇数フィールドではＨＤ₁〜ＨＤ₄、偶数フィールドではＨＤ₁′〜ＨＤ₄′）分の推定式の係数データＷｉが読み出されて推定予測演算回路１２７に供給される。また、バッファメモリ１０９に記憶されているＳＤ信号（５２５ｉ信号）より、第１のタップ選択回路１２１で、作成すべきＨＤ信号（１０５０ｉ信号）を構成する単位画素ブロック内の４画素（注目画素）の周辺に位置する予測タップのデータ（ＳＤ画素データ）が選択的に取り出される。
【０１０６】
推定予測演算回路１２７では、予測タップのデータ（ＳＤ画素データ）ｘｉと、係数メモリ１３４より読み出される４出力画素分の係数データＷｉとから、作成すべきＨＤ信号を構成する単位画素ブロック内の４画素（注目画素）のデータｙ₁〜ｙ₄が演算される（（４）式参照）。そして、この推定予測演算回路１２７より順次出力されるＨＤ信号を構成する単位画素ブロック内の４画素のデータｙ₁〜ｙ₄は正規化演算回路１２８に供給される。
【０１０７】
正規化係数メモリ１３８には上述したようにクラスコードＣＬが読み出しアドレス情報として供給され、この正規化係数メモリ１３８からはクラスコードＣＬに対応した正規化係数Ｓ、つまり推定予測演算回路１２７より出力されるＨＤ画素データｙ₁〜ｙ₄の演算に使用された係数データＷｉに対応した正規化係数Ｓが読み出されて正規化演算回路１２８に供給される。この正規化演算回路１２８では、推定予測演算回路１２７より出力されるＨＤ画素データｙ₁〜ｙ₄がそれぞれ対応する正規化係数Ｓで除算されて正規化される。これにより、係数生成回路１３６で係数データＷｉを求める際の丸め誤差によるデータｙ₁〜ｙ₄のレベル変動が除去される。
【０１０８】
このように正規化演算回路１２８で正規化されて順次出力される単位画素ブロック内の４画素のデータｙ₁′〜ｙ₄′は後処理回路１２９に供給される。この後処理回路１２９では、正規化演算回路１２８より順次供給される単位画素ブロック内の４画素のデータｙ₁′〜ｙ₄′が線順次化され、１０５０ｉ信号のフォーマットで出力される。つまり、後処理回路１２９からは、ＨＤ信号としての１０５０ｉ信号が出力される。
【０１０９】
上述したように、図１に示すテレビ受信機１００では、情報メモリバンク１３５に記憶されている係数種データに基づいて、クラスおよび出力画素の組み合わせ毎に、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉが演算で求められる。このクラスおよび出力画素の組み合わせ毎の推定式の係数データＷｉが係数メモリ１３４に格納される。そして、この係数メモリ１３４より、クラスコードＣＬに対応して読み出される係数データＷｉを用いて推定予測演算回路１２７でＨＤ画素データｙが演算される。
【０１１０】
したがって、ユーザは、値ｈ₁，ｖ₁，ｈ₂，ｖ₂を調整することで、ＨＤ信号によって得られる画像の画質を任意に調整することができる。なおこの場合、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した、クラスおよび出力画素の組み合わせ毎の係数データＷｉをその都度、係数生成回路１３６で生成して使用するものであり、大量の係数データを格納しておくメモリを必要としない。
【０１１１】
また、教師信号から第１の物理的な特徴量としての帯域を変化させた落としフィルタを用いて得られた複数の生徒信号と教師信号との間の学習によって生成される第１の係数種データとしての係数種データｗ1_i0〜ｗ1_i9と、教師信号から第１の物理的な特徴量としての肩特性を変化させた落としフィルタを用いて得られた複数の生徒信号と教師信号との間の学習によって生成される第２の係数種データとしての係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9とに基づいて、推定式の係数データＷｉを生成するものであり、推定式の係数データＷｉとして多様に変化するものを得ることができ、従ってＨＤ信号による画像の画質を多様に調整することができる。
【０１１２】
また、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応した第２の係数種データが情報メモリバンク１３５に存在していいない場合には、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを補間処理によって求めるものであり（（１３）式参照）、情報メモリバンク１３５に格納しておく係数種データを大幅に削減することができ、情報メモリバンク１３５のメモリ容量を大幅に節約できる。また、同様の理由によって、係数種データの算出に伴う学習の数も減らすことができ、開発効率の向上にもつながる。
【０１１３】
なお、上述実施の形態においては、情報メモリバンク１３５に、第２の係数種データとして派生曲面１〜３に対応した係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9が格納されるものを示したが、情報メモリバンク１３５に、さらに多くの派生曲面に対応する係数種データを格納して使用するようにしてもよい。
【０１１４】
また、上述実施の形態においては、情報メモリバンク１３５に、第２の係数種データとして係数種データｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9，ｗ2c_i0〜ｗ2c_i9が格納され、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応した第２の係数種データが情報メモリバンク１３５に存在していない場合には、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを補間処理によって求めるものを示したが、情報メモリバンク１３５に、第２の係数種データとして例えば係数種データｗ2a_i0〜ｗ2a_i9のみが格納され、この係数種データｗ2a_i0〜ｗ2a_i9が値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応していない場合には、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを平行移動処理によって求めることも考えられる。
【０１１５】
この場合、情報メモリバンク１３５には、第１の係数種データとしての係数種データｗ1_i0〜ｗ1_i9と、第２の係数種データとしての係数種データｗ2a_i0〜ｗ2a_i9が格納される。これら係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9の生成方法は上述したと同様である。係数種データｗ1_i0〜ｗ1_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１８に示すような、基底曲面を表すものとなる。また、係数種データｗ2a_i0〜ｗ2a_i9を、（５）式の近似式の係数データｗ_i0〜ｗ_i9に適用した場合、当該（５）式は、図１８に示すような、派生曲面を表すものとなる。
【０１１６】
なお、パラメータｈを固定することで、図１８に示すような係数空間は、図１９に示すように平面で表すことができる。図２０は、落としフィルタの帯域、肩特性と係数データＷｉとの対応関係を示している。左側の係数データＡ〜Ｉが、それぞれ右側のＡ〜Ｉの帯域、肩特性に対応している。
このように情報メモリバンク１３５に係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9が格納される場合、係数生成回路１３６における推定式の係数データＷｉの演算は以下のように行われる。
【０１１７】
図２１は、係数空間における係数データＷｉの指定の流れを示している。まず、値ｈ₁，ｖ₁によって基底曲面上の点を指定し、次に、値ｈ₂，ｖ₂によってこの指定された基底曲面上の点を含む派生曲面上の点を指定する。この指定された派生曲面上の点が、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した、求めるべき係数データＷｉとなる。なお、上述したように、肩特性を変化させる方向を垂直方向として、係数種データｗ2a_i0〜ｗ2a_i9を求めているので、ｈ₁＝ｈ₂である。
【０１１８】
値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した係数データＷｉの求め方を説明する。
まず、情報メモリバンク１３５に存在する係数種データｗ2a_i0〜ｗ2a_i9が、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応している場合について説明する。この場合、係数種データｗ2a_i0〜ｗ2a_i9を、（５）式の係数データｗ_i0〜ｗ_i9に適用すると共に、この（５）式のパラメータｈ，ｖにｈ₂，ｖ₂を代入することで、係数データＷｉを求める。
【０１１９】
次に、情報メモリバンク１３５に存在する係数種データｗ2a_i0〜ｗ2a_i9が、値ｈ₁，ｖ₁によって指定される基底曲面上の点を含む派生曲面（着目派生曲面）に対応していない場合について説明する。この場合、平行移動処理によって、係数データＷｉを求める。図２２を参照して補間処理を説明する。
この場合、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した係数データＷｉをＸとすると、Ｘは（１５）式で算出される。
Ｘ＝Ｗａ＋（Ｗ_v1−Ｗ₆） ・・・（１５）
【０１２０】
ここで、Ｗａは、派生曲面に対応した係数種データｗ2a_i0〜ｗ2a_i9を（５）式の係数データｗ_i0〜ｗ_i9に適用すると共に、この（５）式のパラメータｈ，ｖにｈ₂，ｖaを代入することで、求められる係数データＷｉである。ただし、ｖa＝ｖ₂＋（６−ｖ₁）である。
また、Ｗ_v1，Ｗ₆は、基底曲面に対応した係数種データｗ1_i0〜ｗ1_i9を（５）式の係数データｗ_i0〜ｗ_i9に適用し、この（５）式のパラメータｈ，ｖにそれぞれ「ｈ₁，ｖ₁」、「ｈ₁，６」を代入することで、求められる係数データＷｉである。
【０１２１】
このように、情報メモリバンク１３５に係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9が格納される場合においても、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを良好に得ることができる。また、このように平行移動処理によって推定式の係数データＷｉを求めるものにあっては、上述した補間処理によって推定式の係数データＷｉを求めるものに比べて、情報メモリバンク１３５に格納する係数種データを少なくでき、メモリ容量をさらに節約できる。
また、上述実施の形態においては、肩特性を変化させる方向が垂直方向であるものを示したが、水平方向であっても同様に考えることができる。
【０１２２】
図２３は、上述したテレビ受信機１００の情報メモリバンク１３５に格納される係数種データを生成する係数種データ生成装置１５０の構成を示している。
この係数データ生成装置１５０は、教師信号としてのＨＤ信号（１０５０ｉ信号）が入力される入力端子１５１と、このＨＤ信号に対して水平および垂直の間引き処理を行って、生徒信号としてのＳＤ信号を得るＳＤ信号生成回路１５２とを有している。このＳＤ信号生成回路１５２にはパラメータｈ，ｖが供給される。図示せずも、ＳＤ信号生成回路１５２は、水平および垂直の間引きフィルタからなっている。
【０１２３】
水平間引きフィルタは、間引き処理によって、水平方向の画素数を、ＨＤ信号に対応した数からＳＤ信号に対応した数に減少させる。この水平間引きフィルタは、落としフィルタとしても機能し、供給されるパラメータｈの値によって、水平方向の周波数特性が変化する。すなわち、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を生成する場合、パラメータｈの値が４，５，６，７，８であるとき、図５に示すように、帯域が変化する。
【０１２４】
垂直間引きフィルタは、間引き処理によって、垂直方向の画素数を、ＨＤ信号に対応した数からＳＤ信号に対応した数に減少させる。この垂直間引きフィルタは、落としフィルタとしても機能し、供給されるパラメータｖの値によって、垂直方向の周波数特性が変化する。
【０１２５】
すなわち、係数種データｗ1_i0〜ｗ1_i9を生成する場合、パラメータｖの値が４，５，６，７，８であるとき、図５に示すように、帯域が変化する。また、係数種データｗ2a_i0〜ｗ2a_i9を生成する場合、パラメータｖの値が１，２，３，４，５，６であるとき、図７に示すように、肩特性が変化する。また、係数種データｗ2b_i0〜ｗ2b_i9を生成する場合、パラメータｖの値が０，１，２，３，４，５であるとき、図９に示すように、肩特性が変化する。さらに、係数種データｗ2c_i0〜ｗ2c_i9を生成する場合、パラメータｖの値が−１，０，１，２，３，４であるとき、図１１に示すように、肩特性が変化する。
【０１２６】
また、係数種データ生成装置１５０は、ＳＤ信号生成回路１５２より出力されるＳＤ信号（５２５ｉ信号）より、ＨＤ信号（１０５０ｉ信号）に係る注目画素の周辺に位置する複数のＳＤ画素のデータを選択的に取り出して出力する第１〜第３のタップ選択回路１５３〜１５５を有している。これら第１〜第３のタップ選択回路１５３〜１５５は、上述した画像信号処理部１１０の第１〜第３のタップ選択回路１２１〜１２３と同様に構成される。
【０１２７】
また、係数種データ生成装置１５０は、第２のタップ選択回路１５４で選択的に取り出される空間クラスタップのデータ（ＳＤ画素データ）のレベル分布パターンを検出し、このレベル分布パターンに基づいて空間クラスを検出し、そのクラス情報を出力する空間クラス検出回路１５７を有している。この空間クラス検出回路１５７は、上述した画像信号処理部１１０の空間クラス検出回路１２４と同様に構成される。この空間クラス検出回路１５７からは、空間クラスタップのデータとしての各ＳＤ画素データ毎の再量子化コードｑｉが空間クラスを示すクラス情報として出力される。
【０１２８】
また、係数種データ生成装置１５０は、第３のタップ選択回路１５５で選択的に取り出される動きクラスタップのデータ（ＳＤ画素データ）より、主に動きの程度を表すための動きクラスを検出し、そのクラス情報ＭＶを出力する動きクラス検出回路１５８を有している。この動きクラス検出回路１５８は、上述した画像信号処理部１１０の動きクラス検出回路１２５と同様に構成される。この動きクラス検出回路１５８では、第３のタップ選択回路１５５で選択的に取り出される動きクラスタップのデータ（ＳＤ画素データ）からフレーム間差分が算出され、さらにその差分の絶対値の平均値に対してしきい値処理が行われて動きの指標である動きクラスが検出される。
【０１２９】
また、係数種データ生成装置１５０は、空間クラス検出回路１５７より出力される空間クラスのクラス情報としての再量子化コードｑｉと、動きクラス検出回路１５８より出力される動きクラスのクラス情報ＭＶに基づき、ＨＤ信号（１０５０ｉ信号）に係る注目画素が属するクラスを示すクラスコードＣＬを得るためのクラス合成回路１５９を有している。このクラス合成回路１５９も、上述した画像信号処理部１１０のクラス合成回路１２６と同様に構成される。
【０１３０】
また、係数種データ生成装置１５０は、入力端子１５１に供給されるＨＤ信号から得られる注目画素データとしての各ＨＤ画素データｙと、この各ＨＤ画素データｙにそれぞれ対応して第１のタップ選択回路１５３で選択的に取り出される予測タップのデータ（ＳＤ画素データ）ｘｉと、パラメータｈ，ｖの値と、各ＨＤ画素データｙにそれぞれ対応してクラス合成回路１５９より出力されるクラスコードＣＬとから、クラス毎に、係数種データｗ_i0〜ｗ_i9（ｉ＝１〜ｎ）を得るための正規方程式（（１２）式参照）を生成する正規方程式生成部１６０を有している。
【０１３１】
この場合、１個のＨＤ画素データｙとそれに対応するｎ個の予測タップのデータ（ＳＤ画素データ）ｘｉとの組み合わせで学習データが生成されるが、パラメータｈ，ｖの値の変化に対応してＳＤ信号生成回路１５２における水平および垂直の周波数特性が可変され、複数のＳＤ信号が順次生成されていく。これにより、正規方程式生成部１６０では、パラメータｈ，ｖの値が異なる多くの学習データが登録された正規方程式が生成され、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9をそれぞれ求めることが可能となる。
【０１３２】
またこの場合、１個のＨＤ画素データｙとそれに対応するｎ個の予測タップのデータ（ＳＤ画素データ）ｘｉとの組み合わせで学習データが生成されるが、正規方程式生成部１６０では、出力画素（図２のＨＤ₁〜ＨＤ₄、図３のＨＤ１′〜ＨＤ₄′参照）毎に、正規方程式が生成される。例えば、ＨＤ₁に対応した正規方程式は、中心予測タップに対するずれ値がＨＤ₁と同じ関係にあるＨＤ画素データｙから構成される学習データから生成される。
【０１３３】
また、係数種データ生成装置１５０は、正規方程式生成部１６０で、クラスおよび出力画素の組み合わせ毎に生成された正規方程式のデータが供給され、当該正規方程式を解いて、クラスおよび出力画素の組み合わせ毎に、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を求める係数種データ決定部１６１と、この求められた係数種データを記憶する係数種メモリ１６２とを有している。係数種データ決定部１６１では、正規方程式が例えば掃き出し法などによって解かれて、係数種データが求められる。
【０１３４】
図２３に示す係数種データ生成装置１５０の動作を説明する。
入力端子１５１には教師信号としてのＨＤ信号（１０５０ｉ信号）が供給され、そしてこのＨＤ信号に対してＳＤ信号生成回路１５２で水平および垂直の間引き処理が行われて生徒信号としてのＳＤ信号（５２５ｉ信号）が生成される。
【０１３５】
このＳＤ信号（５２５ｉ信号）より、第２のタップ選択回路１５４で、ＨＤ信号（１０５０ｉ信号）に係る注目画素の周辺に位置する空間クラスタップのデータ（ＳＤ画素データ）が選択的に取り出される。この第２のタップ選択回路１５４で選択的に取り出される空間クラスタップのデータ（ＳＤ画素データ）は空間クラス検出回路１５７に供給される。この空間クラス検出回路１５７では、空間クラスタップのデータとしての各ＳＤ画素データに対してＡＤＲＣ処理が施されて空間クラス（主に空間内の波形表現のためのクラス分類）のクラス情報としての再量子化コードｑｉが得られる（（１）式参照）。
【０１３６】
また、ＳＤ信号生成回路１５２で生成されたＳＤ信号より、第３のタップ選択回路１５５で、ＨＤ信号に係る注目画素の周辺に位置する動きクラスタップのデータ（ＳＤ画素データ）が選択的に取り出される。この第３のタップ選択回路１５５で選択的に取り出される動きクラスタップのデータ（ＳＤ画素データ）は動きクラス検出回路１５８に供給される。この動きクラス検出回路１５８では、動きクラスタップのデータとしての各ＳＤ画素データより動きクラス（主に動きの程度を表すためのクラス分類）のクラス情報ＭＶが得られる。
【０１３７】
このクラス情報ＭＶと上述した再量子化コードｑｉはクラス合成回路１５９に供給される。このクラス合成回路１５９では、これらクラス情報ＭＶと再量子化コードｑｉとから、ＨＤ信号（１０５０ｉ信号）に係る注目画素が属するクラスを示すクラスコードＣＬが得られる（（３）式参照）。
【０１３８】
また、ＳＤ信号生成回路１５２で生成されるＳＤ信号より、第１のタップ選択回路１５３で、ＨＤ信号に係る注目画素の周辺に位置する予測タップのデータ（ＳＤ画素データ）が選択的に取り出される。
【０１３９】
そして、入力端子１５１に供給されるＨＤ信号より得られる注目画素データとしての各ＨＤ画素データｙと、この各ＨＤ画素データｙにそれぞれ対応して第１のタップ選択回路１５３で選択的に取り出される予測タップのデータ（ＳＤ画素データ）ｘｉと、パラメータｈ，ｖの値と、各ＨＤ画素データｙにそれぞれ対応してクラス合成回路１５９より出力されるクラスコードＣＬとから、正規方程式生成部１６０では、クラスおよび出力画素の組み合わせ毎に、係数種データｗ_i0〜ｗ_i9を得るための正規方程式（（１２）式参照）が個別に生成される。
【０１４０】
ここで、係数種データｗ1_i0〜ｗ1_i9を生成する場合、パラメータｈ，ｖが４，５，６，７，８のように順次変更され、そのとき垂直、水平の間引きフィルタの帯域が図５に示すように変化するようにされるため、正規方程式生成部１６０で生成される正規方程式は係数種データｗ1_i0〜ｗ1_i9を得るためのものとなる。
【０１４１】
また、係数種データｗ2a_i0〜ｗ2a_i9を生成する場合、パラメータｈが４，５，６，７，８のように順次変更され、そのとき水平間引きフィルタの帯域が図５に示すように変化するようにされると共に、パラメータｖが１，２，３，４，５，６のように順次変更され、そのとき垂直間引きフィルタの肩特性が図７に示すように変化するようにされる。そのため、正規方程式生成部１６０で生成される正規方程式は係数種データｗ2a_i0〜ｗ2a_i9を得るためのものとなる。
【０１４２】
また、係数種データｗ2b_i0〜ｗ2b_i9を生成する場合、パラメータｈが４，５，６，７，８のように順次変更され、そのとき水平間引きフィルタの帯域が図５に示すように変化するようにされると共に、パラメータｖが０，１，２，３，４，５のように順次変更され、そのとき垂直間引きフィルタの肩特性が図９に示すように変化するようにされる。そのため、正規方程式生成部１６０で生成される正規方程式は係数種データｗ2b_i0〜ｗ2b_i9を得るためのものとなる。
【０１４３】
また、係数種データｗ2c_i0〜ｗ2c_i9を生成する場合、パラメータｈが４，５，６，７，８のように順次変更され、そのとき水平間引きフィルタの帯域が図５に示すように変化するようにされると共に、パラメータｖが−１，０，１，２，３，４のように順次変更され、そのとき垂直間引きフィルタの肩特性が図１１に示すように変化するようにされる。そのため、正規方程式生成部１６０で生成される正規方程式は係数種データｗ2c_i0〜ｗ2c_i9を得るためのものとなる。
【０１４４】
そして、係数データ決定部１６１で各正規方程式が解かれ、クラスおよび出力画素の組み合わせ毎に、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9が求められ、それらの係数種データは係数種メモリ１６２に記憶される。
【０１４５】
このように、図２３に示す係数種データ生成装置１５０においては、図１の画像信号処理部１１０の情報メモリバンク１３５に記憶される、クラスおよび出力画素（ＨＤ₁〜ＨＤ₄，ＨＤ₁′〜ＨＤ₄′）の組み合わせ毎の、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を生成することができる。
なお、図１の画像信号処理部１１０における処理を、例えば図２４に示すような画像信号処理装置３００によって、ソフトウェアで実現することも可能である。
【０１４６】
まず、図２４に示す画像信号処理装置３００について説明する。この画像信号処理装置３００は、装置全体の動作を制御するＣＰＵ３０１と、このＣＰＵ３０１の処理プログラム、さらに推定式の係数データＷｉを生成するために使用される係数種データ（図１の画像信号処理部１１０の情報メモリバンク１３５に予め蓄えられている情報と同じ）等が格納されたＲＯＭ（read only memory）３０２と、ＣＰＵ３０１の作業領域を構成するＲＡＭ（random access memory）３０３とを有している。これらＣＰＵ３０１、ＲＯＭ３０２およびＲＡＭ３０３は、それぞれバス３０４に接続されている。
【０１４７】
また、画像信号処理装置３００は、外部記憶装置としてのハードディスクドライブ（ＨＤＤ）３０５と、フロッピー（Ｒ）ディスク３０６をドライブするフロッピー（Ｒ）ディスクドライブ（ＦＤＤ）３０７とを有している。これらドライブ３０５，３０７は、それぞれバス３０４に接続されている。
【０１４８】
また、画像信号処理装置３００は、インターネット等の通信網４００に有線または無線で接続する通信部３０８を有している。この通信部３０８は、インタフェース３０９を介してバス３０４に接続されている。
【０１４９】
また、画像信号処理装置３００は、ユーザインタフェース部を備えている。このユーザインタフェース部は、リモコン送信機２００からのリモコン信号ＲＭを受信するリモコン信号受信回路３１０と、ＬＣＤ（liquid crystal display）等からなるディスプレイ３１１とを有している。受信回路３１０はインタフェース３１２を介してバス３０４に接続され、同様にディスプレイ３１１はインタフェース３１３を介してバス３０４に接続されている。
【０１５０】
また、画像信号処理装置３００は、ＳＤ信号を入力するための入力端子３１４と、ＨＤ信号を出力するための出力端子３１５とを有している。入力端子３１４はインタフェース３１６を介してバス３０４に接続され、同様に出力端子３１５はインタフェース３１７を介してバス３０４に接続される。
【０１５１】
ここで、上述したようにＲＯＭ３０２に処理プログラム等を予め格納しておく代わりに、例えばインターネットなどの通信網４００より通信部３０８を介してダウンロードし、ハードディスクやＲＡＭ３０３に蓄積して使用することもできる。また、これら処理プログラム、係数種データ等をフロッピー（Ｒ）ディスク３０６で提供するようにしてもよい。
【０１５２】
また、処理すべきＳＤ信号を入力端子３１４より入力する代わりに、予めハードディスクに記録しておき、あるいはインターネットなどの通信網４００より通信部３０８を介してダウンロードしてもよい。また、処理後のＨＤ信号を出力端子３１５に出力する代わり、あるいはそれと並行してディスプレイ３１１に供給して画像表示をしたり、さらにはハードディスクに格納したり、通信部３０８を介してインターネットなどの通信網４００に送出するようにしてもよい。
【０１５３】
図２５のフローチャートを参照して、図２４に示す画像信号処理装置３００における、ＳＤ信号よりＨＤ信号を得るため処理手順を説明する。
まず、ステップＳＴ１で、処理を開始し、ステップＳＴ２で、ＳＤ画素データをフレーム単位またはフィールド単位で入力する。このＳＤ画素データが入力端子３１４より入力される場合には、このＳＤ画素データをＲＡＭ３０３に一時的に格納する。また、このＳＤ画素データがハードディスクに記録されている場合には、ハードディスクドライブ３０７でこのＳＤ画素データを読み出し、ＲＡＭ３０３に一時的に格納する。そして、ステップＳＴ３で、入力ＳＤ画素データの全フレームまたは全フィールドの処理が終わっているか否かを判定する。処理が終わっているときは、ステップＳＴ４で、処理を終了する。一方、処理が終わっていないときは、ステップＳＴ５に進む。
【０１５４】
このステップＳＴ５では、ユーザがリモコン送信機２００を操作して入力した画質指定値、つまりｈ₁，ｖ₁，ｈ₂，ｖ₂を例えばＲＡＭ３０３より読み込む。そして、ステップＳＴ６で、例えばＲＯＭ３０２から、推定式の係数データＷｉを生成するために使用する係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を読み込む。
【０１５５】
次に、ステップＳＴ７で、ステップＳＴ５で読み込んだ値ｈ₁，ｖ₁，ｈ₂，ｖ₂およびステップＳＴ６で取得された係数種データとから、クラスおよび出力画素の組み合わせ毎に、値ｈ₁，ｖ₁，ｈ₂，ｖ₂に対応した推定式の係数データＷｉを生成する。
【０１５６】
次に、ステップＳＴ８で、ステップＳＴ２で入力されたＳＤ画素データより、生成すべき各ＨＤ画素データに対応して、クラスタップおよび予測タップの画素データを取得する。そして、ステップＳＴ９で、入力されたＳＤ画素データの全領域においてＨＤ画素データを得る処理が終了したか否かを判定する。終了しているときは、ステップＳＴ２に戻り、次のフレームまたはフィールドのＳＤ画素データの入力処理に移る。一方、処理が終了していないときは、ステップＳＴ１０に進む。
【０１５７】
このステップＳＴ１０では、ステップＳＴ９で取得されたクラスタップのＳＤ画素データからクラスコードＣＬを生成する。そして、ステップＳＴ１１で、そのクラスコードＣＬに対応した係数データと予測タップのＳＤ画素データを使用して、推定式により、ＨＤ画素データを生成し、その後にステップＳＴ８に戻って、上述したと同様の処理を繰り返す。
【０１５８】
このように、図２５に示すフローチャートに沿って処理をすることで、入力されたＳＤ信号を構成するＳＤ画素データを処理して、ＨＤ信号を構成するＨＤ画素データを得ることができる。上述したように、このように処理して得られたＨＤ信号は出力端子３１５に出力されたり、ディスプレイ３１１に供給されてそれによる画像が表示されたり、さらにはハードディスクドライブ３０５に供給されてハードディスクに記録されたりする。
また、処理装置の図示は省略するが、図２３の係数種データ生成装置１５０における処理を、ソフトウェアで実現することも可能である。
【０１５９】
図２６のフローチャートを参照して、推定式の係数データＷｉを生成するために使用する係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を生成する処理手順を説明する。
まず、ステップＳＴ２１で、処理を開始し、ステップＳＴ２２で、学習に使われるＳＤ画素データを生成する際に使用する落としフィルタの水平、垂直の周波数特性を選択する。そして、ステップＳＴ２３で、水平、垂直の全周波数特性に対して学習が終わったか否かを判定する。全周波数特性に対して学習が終わっていないときは、ステップＳＴ２４に進む。
【０１６０】
このステップＳＴ２４では、既知のＨＤ画素データをフレーム単位またはフィールド単位で入力する。そして、ステップＳＴ２５で、全てのＨＤ画素データについて処理が終了したか否かを判定する。終了したときは、ステップＳＴ２２に戻って、次の周波数特性を選択して、上述したと同様の処理を繰り返す。一方、終了していないときは、ステップＳＴ２６に進む。
【０１６１】
このステップＳＴ２６では、ステップＳＴ２４で入力されたＨＤ画素データより、ステップＳＴ２２で選択された水平、垂直の周波数特性の落としフィルタを使用してＳＤ画素データを生成する。そして、ステップＳＴ２７で、ステップＳＴ２６で生成されたＳＤ画素データより、ステップＳＴ２４で入力された各ＨＤ画素データに対応してクラスタップおよび予測タップの画素データを取得する。
【０１６２】
そして、ステップＳＴ２８で、生成されたＳＤ画素データの全領域において学習処理を終了しているか否かを判定する。学習処理を終了しているときは、ステップＳＴ２４に戻って、次のＨＤ画素データの入力を行って、上述したと同様の処理を繰り返し、一方、学習処理を終了していないときは、ステップＳＴ２９に進む。
【０１６３】
このステップＳＴ２９では、ステップＳＴ２７で取得されたクラスタップのＳＤ画素データからクラスコードＣＬを生成する。そして、ステップＳＴ３０で、係数種データｗ_i0〜ｗ_i9を得るための正規方程式（（１２）式参照）を生成する。ここでは、クラスおよび出力画素の組み合わせ毎に、正規方程式が個別に生成される。その後に、ステップＳＴ２７に戻る。
【０１６４】
また、ステップＳＴ２３で、全ての周波数特性に対して学習が終わったときは、ステップＳＴ３１に進む。このステップＳＴ３１では、各正規方程式を掃き出し法等で解くことによって、クラスおよび出力画素の組み合わせ毎の係数種データｗ_i0〜ｗ_i9を算出し、ステップＳＴ３２で、その係数種データｗ_i0〜ｗ_i9をメモリに保存する。
【０１６５】
ここで、ステップＳＴ２２で係数種ｗ1_i0〜ｗ1_i9を得るための周波数特性を選択することで、ステップＳＴ３０では係数種データｗ_i0〜ｗ_i9として係数種ｗ1_i0〜ｗ1_i9を得るための正規方程式が生成され、ステップＳＴ３１で算出される係数種データｗ_i0〜ｗ_i9はｗ1_i0〜ｗ1_i9となる。以下同様にし、ステップＳＴ２２で選択される周波数特性を変更することで、ステップＳＴ３１で算出される係数種データｗ_i0〜ｗ_i9はｗ2a_i0〜ｗ2a_i9、ｗ2b_i0〜ｗ2b_i9またはｗ2c_i0〜ｗ2c_i9となる。
【０１６６】
次に、ステップＳＴ３３で、係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9の全てを算出したか否かを判定する。全てを算出していないときは、ステップＳＴ２２に戻って、次の係数種データを算出するための処理に移る。一方、全てを算出したときは、ステップＳＴ３４で、処理を終了する。
【０１６７】
このように、図２６に示すフローチャートに沿って処理をすることで、図２３に示す係数種データ生成装置１５０と同様の手法によって、図１の画像信号処理部１１０の情報メモリバンク１３５に記憶される係数種データｗ1_i0〜ｗ1_i9，ｗ2a_i0〜ｗ2a_i9，ｗ2b_i0〜ｗ2b_i9およびｗ2c_i0〜ｗ2c_i9を生成することができる。
【０１６８】
なお、上述実施の形態においては、落としフィルタの周波数特性を変化させる物理的な特徴量として、帯域および肩特性の２種類を使用する例を示したが、３種類以上の物理的な特徴量を使用することもできる。この場合、第３以降の物理的な特徴量によって、それぞれ他の物理的な特徴量の所定値に対応した周波数特性よりこの周波数特性を変化させた落としフィルタを用いて得られた生徒信号を用いて、第３以降の係数種データの算出が行われる。
【０１６９】
したがって、第３以降の係数種データに対応した派生曲面は、基底曲面や他の派生曲面に必ず交わる。例えば、第３の係数種データを考えると、この第３の係数種データに対応した派生曲面は、図２７Ａに示すように、第１の係数種データに対応した基底曲面と交わるか、あるいは図２７Ｂに示すように、第２の係数種データに対応した派生曲面と交わる。
【０１７０】
図２７Ａの場合、第３の物理的な特徴量がｖ＝６に対応した値であるときの落としフィルタの周波数特性は、第１の物理的な特徴量（例えば帯域）がｖ＝６に対応した値であるときの落としフィルタの周波数特性と同じくなる。また、図２７Ｂの場合、第３の物理的な特徴量がｖ＝１に対応した値であるときの落としフィルタの周波数特性は、第２の物理的な特徴量（肩特性）がｖ＝１に対応した値であるときの落としフィルタの周波数特性と同じくなる。
【０１７１】
ユーザは、使用する物理的な特徴量の数に対応した画質の調整軸をもつことになる。この場合、図２７Ａに示すように基底曲面▲１▼に派生曲面▲２▼，▲３▼が交わる場合と、図２７Ｂに示すように基底曲面▲１▼に派生曲面▲２▼が交わり、この派生曲面▲２▼に派生曲面▲３▼が交わる場合とでは、係数データＷｉの指定の仕方が異なってくる。
【０１７２】
図２７Ａに示すような場合、図２８に示すように、まず、基底曲面▲１▼上の点を指定し、この指定された点を含む派生曲面▲２▼′または▲３▼′上の点を指定する。つまり、この場合には、派生曲面▲２▼または▲３▼に対応した調整軸のいずれかを選択的に使用して調整することとなる。
【０１７３】
図２７Ｂに示すような場合、図２９に示すように、まず、基底曲面▲１▼上の点を指定し、次に、この指定された点を含む派生曲面▲２▼′上の点を指定し、さらにこの指定された点を含む派生曲面▲３▼′上の点を指定する。つまり、この場合には、派生曲面▲２▼，▲３▼に対応した調整軸の双方を使用して調整することとなる。
【０１７４】
このように、曲面の空間的なつながりを利用して、様々な調整軸で画質を調整でき、ユーザの好みに応じた任意の画質を得ることができる。このように、３種類以上の物理的な特徴量を使用する場合にも、上述した補間処理や平行移動処理等によって、指定された係数データＷｉを生成して、推定予測演算回路１２７で使用することができる。
【０１７５】
また、上述実施の形態において、情報メモリバンク１３５には、出力画素（ＨＤ₁〜ＨＤ₄、ＨＤ₁′〜ＨＤ₄′）毎に、推定式の係数データＷｉを生成するための係数種データが記憶されている。ＳＤ信号（５２５ｉ信号）の画素に対する出力画素の位相を変更することで、１０５０ｉ信号以外のＨＤ信号も同様にして得ることができる。
また、上述実施の形態においては、クラス分けをするものを示したが、この発明はクラス分けをしないものにも同様に適用できることは勿論である。
【０１７６】
また、上述実施の形態においては、ＨＤ信号を生成する際の推定式として線形一次方程式を使用したものを挙げたが、これに限定されるものではなく、例えば推定式として高次方程式を使用するものであってもよい。
また、上述実地の形態においては、情報信号が画像信号である場合を示したが、この発明はこれに限定されない。例えば、情報信号が音声信号である場合にも、この発明を同様に適用することができる。
【０１７７】
【発明の効果】
この発明によれば、第１の情報信号を第２の情報信号に変換する際に使用される推定式の係数データを、教師信号から複数の物理的な特徴量によってそれぞれ周波数特性を変化させた落としフィルタを用いて得られた複数種類の生徒信号と教師信号との間の学習によって生成される複数種類の係数種データに基づいて生成するものであり、係数データを格納しておくメモリの容量増加を招くことなく、パラメータの値の変化に対応した係数データを得ることができ、従ってパラメータの値を調整して第２の情報信号による出力の質を任意に調整することができ、また多様に変化する係数データを得ることができ、従って第２の情報信号による出力の質を多様に調整することができる。
【図面の簡単な説明】
【図１】実施の形態としてのテレビ受信機の構成を示すブロック図である。
【図２】ＨＤ信号（１０５０ｉ信号）の単位画素ブロック内の４画素の中心予測タップからの位相ずれ（奇数フィールド）を示す図である。
【図３】ＨＤ信号（１０５０ｉ信号）の単位画素ブロック内の４画素の中心予測タップからの位相ずれ（偶数フィールド）を示す図である。
【図４】係数種データｗ1_i0〜ｗ1_i9の生成方法を示す図である。
【図５】パラメータｈ，ｖを４，５，６，７，８としたときの落としフィルタにおける水平、垂直の周波数特性（帯域変化）を示す図である。
【図６】係数種データｗ2a_i0〜ｗ2a_i9の生成方法を示す図である。
【図７】パラメータｖを１，２，３，４，５，６としたときの落としフィルタにおける垂直周波数特性（肩特性変化）を示す図である。
【図８】係数種データｗ2b_i0〜ｗ2b_i9の生成方法を示す図である。
【図９】パラメータｖを０，１，２，３，４，５としたときの落としフィルタにおける垂直周波数特性（肩特性変化）を示す図である。
【図１０】係数種データｗ2c_i0〜ｗ2c_i9の生成方法を示す図である。
【図１１】パラメータｖを−１，０，１，２，３，４としたときの落としフィルタにおける垂直周波数特性（肩特性変化）を示す図である。
【図１２】基底曲面、派生曲面を含む係数空間の模式図である。
【図１３】基底曲面、派生曲面を含む係数空間（平面）の模式図である。
【図１４】落としフィルタの帯域、肩特性と係数データＷｉとの対応関係を示す図である。
【図１５】係数空間における係数データＷｉの指定の流れを示す図である。
【図１６】着目派生曲面が派生曲面２，３の間に存在する場合を示す図である。
【図１７】補間処理を説明するための図である。
【図１８】基底曲面、派生曲面を含む係数空間の模式図である。
【図１９】基底曲面、派生曲面を含む係数空間（平面）の模式図である。
【図２０】落としフィルタの帯域、肩特性と係数データＷｉとの対応関係を示す図である。
【図２１】係数空間における係数データＷｉの指定の流れを示す図である。
【図２２】平行移動処理を説明するための図である。
【図２３】係数種データ生成装置の構成例を示すブロック図である。
【図２４】ソフトウェアで実現するための画像信号処理装置の構成例を示すブロック図である。
【図２５】画像信号処理を示すフローチャートである。
【図２６】係数種データ生成処理を示すフローチャートである。
【図２７】第３の係数種データを考えた場合の係数空間の模式図である。
【図２８】係数データの指定の仕方を説明するための図である。
【図２９】係数データの指定の仕方を説明するための図である。
【図３０】５２５ｉ信号と１０５０ｉ信号の画素位置関係を示す図である。
【符号の説明】
１００・・・テレビ受信機、１０１・・・システムコントローラ、１０２・・・リモコン信号受信回路、１０５・・・受信アンテナ、１０６・・・チューナ、１１０・・・画像信号処理部、１１１・・・ディスプレイ部、１１２・・・ＯＳＤ回路、１２１・・・第１のタップ選択回路、１２２・・・第２のタップ選択回路、１２３・・・第３のタップ選択回路、１２４・・・空間クラス検出回路、１２５・・・動きクラス検出回路、１２６・・・クラス合成回路、１２７・・・推定予測演算回路、１２８・・・正規化演算回路、１２９・・・後処理回路、１３４・・・係数メモリ、１３５・・・情報メモリバンク、１３６・・・係数生成回路、１３７・・・正規化係数演算部、１３８・・・正規化係数メモリ、１５０・・・係数種データ生成装置，１５１・・・入力端子、１５２・・・ＳＤ信号生成回路、１５３・・・第１のタップ選択回路、１５４・・・第２のタップ選択回路、１５５・・・第３のタップ選択回路、１５７・・・空間クラス検出回路、１５８・・・動きクラス検出回路、１５９・・・クラス合成回路、１６０・・・正規方程式生成部、１６１・・・係数種データ決定部、１６２・・・係数種メモリ、２００・・・リモコン送信機、３００・・・画像信号処理装置[0001]
BACKGROUND OF THE INVENTION
The present invention can be applied, for example, when converting an NTSC video signal to a high-definition video signal, a coefficient data generating apparatus and generating method, an information signal processing apparatus and processing method using the coefficient data generating apparatus, and a method for use therein. Coefficient seed data generation device and generation method To the law Related. Specifically, the coefficient data of the estimation formula used when converting the first information signal into the second information signal is a drop filter in which the frequency characteristics are changed from the teacher signal according to a plurality of physical feature amounts, respectively. Generating based on a plurality of types of coefficient seed data generated by learning between a plurality of types of student signals and teacher signals obtained by using the same leads to an increase in the capacity of the memory for storing the coefficient data The present invention relates to a coefficient data generation device or the like that can obtain coefficient data corresponding to changes in parameter values and can obtain coefficient data that varies in various ways.
[0002]
[Prior art]
Conventionally, for example, a format conversion for converting an SD (Standard Definition) signal called a 525i signal into an HD (High Definition) signal called a 1050i signal has been proposed. The 525i signal means an interlaced image signal having 525 lines, and the 1050i signal means an interlaced image signal having 1050 lines.
[0003]
FIG. 30 shows the pixel position relationship between the 525i signal and the 1050i signal. Here, a large dot is a pixel of a 525i signal, and a small dot is a pixel of a 1050i signal. In addition, pixel positions in odd fields are indicated by solid lines, and pixel positions in even fields are indicated by broken lines. When converting a 525i signal to a 1050i signal, it is necessary to obtain four pixels of the 1050i signal corresponding to one pixel of the 525i signal in each of the odd and even fields.
[0004]
Conventionally, when the pixel data of the 1050i signal is obtained from the pixel data of the 525i signal in order to perform the format conversion as described above, the coefficient data of the estimation formula corresponding to the phase of each pixel of the 1050i signal with respect to the pixel of the 525i signal is obtained. It has been proposed that pixel data of a 1050i signal is obtained by an estimation formula using the coefficient data stored in a memory.
[0005]
[Problems to be solved by the invention]
As described above, in the case of obtaining the pixel data of the 1050i signal by the estimation formula, the resolution of the image by the 1050i signal is fixed, and it is desired according to the image content or the like like the conventional adjustment of contrast, sharpness, etc. The resolution could not be achieved. It is conceivable that coefficient data corresponding to a plurality of resolutions is prepared in order for the user to arbitrarily adjust to a desired resolution. However, the capacity of the memory for storing the coefficient data is increased, and the cost of the apparatus is increased. Leads to.
[0006]
According to the present invention, coefficient data corresponding to a change in parameter value can be obtained without causing an increase in the capacity of a memory for storing coefficient data, and coefficient data changing in various ways can be obtained. An object of the present invention is to provide a coefficient data generation device and the like.
[0007]
[Means for Solving the Problems]
The coefficient data generation device according to the present invention is used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data, and is focused on the second information signal. Is a coefficient data generation device that generates coefficient data of an estimation formula for calculating a plurality of pieces of information data extracted from the first information signal, and a teacher signal corresponding to the second information signal; The learning signal is generated by learning with a plurality of student signals corresponding to the first information signal obtained by using the drop filter whose frequency characteristic is changed by the first physical feature amount. , Corresponding to the first coefficient seed data, which is the coefficient data of the generation formula for generating the coefficient data of the estimation formula, including the first parameter indicating the first physical feature quantity, and the second information signal Teacher signal From the teacher signal, the second to Nth (N is an integer greater than or equal to 2) physical feature values change the frequency characteristics from frequency characteristics corresponding to predetermined values of other physical feature values. 2nd to 2nd indicating the second to Nth physical feature amounts respectively generated by learning with a plurality of student signals corresponding to the first information signal obtained using the dropped filter. Storage means for storing second to Nth coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of the estimation formula including N parameters, a value of the first parameter, and second to second A parameter setting means for setting the value of the N parameter, and a first coefficient set by the parameter setting means based on the first coefficient seed data and the second to Nth coefficient seed data stored in the storage means. Parameters and 2nd to 2nd Corresponding to the value of the parameter, in which and a coefficient data generation means for generating coefficient data of said estimated equation.
[0008]
The coefficient data generation method according to the present invention is used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data, and relates to the second information signal. A coefficient data generation method for generating coefficient data of an estimation formula for calculating information data of a point of interest from a plurality of information data extracted from a first information signal, the teacher corresponding to a second information signal Generated by learning between the signal and a plurality of student signals corresponding to the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by the first physical feature amount The first coefficient seed data that is the coefficient data of the generation formula for generating the coefficient data of the estimation formula including the first parameter indicating the first physical feature value is obtained from the storage unit When A teacher signal corresponding to the second information signal, and a predetermined value of another physical feature amount based on the second to Nth (N is an integer of 2 or more) physical feature amounts from the teacher signal. Are generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from the frequency characteristic corresponding to. A step of acquiring from the storage unit second to Nth coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of the estimation formula, including second to Nth parameters indicating physical feature amounts Setting the first parameter value and the second to Nth parameter values, and the obtained first coefficient seed data and the second to Nth coefficient seed data. The value of the first parameter and the Corresponding to the value of the parameter th to N, in which and a step of generating the coefficient data of said estimated equation.
[0010]
In the present invention, the first coefficient seed data and the second to Nth coefficient seed data are stored in the storage means.
The first coefficient seed data is obtained by using a teacher signal corresponding to the second information signal and a first filter obtained from the teacher signal using a drop filter whose frequency characteristic is changed by the first physical feature amount. In order to generate coefficient data of the estimation formula including a first parameter indicating a first physical feature amount, which is generated by learning with a plurality of student signals corresponding to the information signal of Is the coefficient data of the generation formula.
[0011]
Similarly, the second to N-th coefficient seed data includes a teacher signal corresponding to the second information signal and another physical feature value from the teacher signal according to the second to N-th physical feature amounts. Generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter whose frequency characteristic is changed from a frequency characteristic corresponding to a predetermined value of a characteristic amount. The coefficient data of the generation formula includes the second to Nth parameters indicating the second to Nth physical feature amounts, respectively, and generates the coefficient data of the estimation formula.
[0012]
For example, the storage means stores second coefficient seed data in addition to the first coefficient seed data. Further, for example, the storage means stores second and third coefficient seed data in addition to the first coefficient seed data. The third coefficient seed data includes a teacher signal corresponding to the second information signal and a frequency characteristic corresponding to a predetermined value of the first parameter or a predetermined value of the second parameter from the teacher signal. 3 is generated by learning with a plurality of student signals corresponding to the first information signal obtained by using the drop filter whose frequency characteristic is changed by the physical feature amount of 3.
[0013]
Based on the first coefficient seed data stored in the storage means and the second to Nth coefficient seed data, the values of the first parameter and the second to Nth parameters set by the parameter setting means are set. Corresponding estimation formula coefficient data is generated.
[0014]
In this way, a drop filter in which the coefficient data of the estimation formula used when converting the first information signal into the second information signal is changed in frequency characteristics from the teacher signal according to a plurality of physical feature amounts, respectively. Is generated on the basis of a plurality of types of coefficient seed data generated by learning between a plurality of types of student signals and teacher signals obtained by using, thereby increasing the capacity of a memory for storing coefficient data Therefore, coefficient data corresponding to the change of the parameter value can be obtained, and therefore, the parameter value can be adjusted to arbitrarily adjust the output quality by the second information signal, and various changes can be made. The coefficient data to be obtained can be obtained, and therefore the quality of the output by the second information signal can be variously adjusted.
[0015]
An information signal processing apparatus according to the present invention is an information signal processing apparatus that converts a first information signal made up of a plurality of information data into a second information signal made up of a plurality of information data. Between the corresponding teacher signal and a plurality of student signals corresponding to the first information signal obtained from the teacher signal by using a drop filter whose frequency characteristic is changed by the first physical feature amount First coefficient seed data which is coefficient data of a generation formula for generating coefficient data of an estimation formula including a first parameter indicating a first physical feature value generated by learning; The teacher signal corresponding to the information signal and the second to Nth (N is an integer of 2 or more) physical feature values from the teacher signal correspond to predetermined values of other physical feature values, respectively. This frequency characteristic is better than the frequency characteristic Second to Nth physical feature amounts respectively generated by learning with a plurality of student signals corresponding to the first information signal obtained by using the reduced filtering filter. Storage means for storing second to Nth coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of the estimation formula including the Nth parameter, a value of the first parameter, and second to second The parameter setting means for setting the value of the Nth parameter, the first coefficient seed data stored in the storage means, and the first set by the parameter setting means based on the second to Nth coefficient seed data. Coefficient data generating means for generating coefficient data of the estimation equation corresponding to the parameter values of the first and second to Nth parameters, and the vicinity of the point of interest relating to the second information signal from the first information signal Multiple located in From the first data selection means for selecting the first information data, the coefficient data generated by the coefficient data generation means, and the plurality of first information data selected by the first data selection means, the above estimation formula And calculating means for calculating the information data of the attention point using.
[0016]
The information signal processing method according to the present invention is an information signal processing method for converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data, wherein the second information signal A teacher signal corresponding to the signal and a plurality of student signals corresponding to the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by the first physical feature amount Storing first coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of an estimation formula, including a first parameter indicating a first physical feature value, generated by learning between , The teacher signal corresponding to the second information signal, and the second to N-th (N is an integer of 2 or more) physical feature values from the teacher signal, respectively, Corresponding to predetermined values of various features Are generated by learning with a plurality of student signals corresponding to the first information signal obtained by using the drop filter in which the frequency characteristic is changed from the frequency characteristic. Obtaining from the storage unit the second to Nth coefficient seed data, which is the coefficient data of the generation formula for generating the coefficient data of the estimation formula, including the second to Nth parameters indicating various feature quantities; The first parameter value and the second to Nth parameter values are set, and the set first coefficient seed data and the second to Nth coefficient seed data are set based on the first coefficient seed data and the second to Nth coefficient seed data. Generating coefficient data of the estimation formula corresponding to the parameter value and the second to Nth parameter values, and located around the point of interest related to the second information signal from the first information signal Multiple information data Selecting a, from the generated coefficient data and the selected plurality of information data, and comprising the steps that can calculate the information data of said target point by using the estimation equation.
[0018]
In the present invention, the first coefficient seed data and the second to Nth coefficient seed data are stored in the storage means.
The first coefficient seed data is obtained by using a teacher signal corresponding to the second information signal and a first filter obtained from the teacher signal using a drop filter whose frequency characteristic is changed by the first physical feature amount. In order to generate coefficient data of the estimation formula including a first parameter indicating a first physical feature amount, which is generated by learning with a plurality of student signals corresponding to the information signal of Is the coefficient data of the generation formula.
[0019]
Similarly, the second to N-th coefficient seed data includes a teacher signal corresponding to the second information signal and another physical feature value from the teacher signal according to the second to N-th physical feature amounts. Generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter whose frequency characteristic is changed from a frequency characteristic corresponding to a predetermined value of a characteristic amount. The coefficient data of the generation formula includes the second to Nth parameters indicating the second to Nth physical feature amounts, respectively, and generates the coefficient data of the estimation formula.
[0020]
For example, the storage means stores second coefficient seed data in addition to the first coefficient seed data. Further, for example, the storage means stores second and third coefficient seed data in addition to the first coefficient seed data. The third coefficient seed data includes a teacher signal corresponding to the second information signal and a frequency characteristic corresponding to a predetermined value of the first parameter or a predetermined value of the second parameter from the teacher signal. 3 is generated by learning with a plurality of student signals corresponding to the first information signal obtained by using the drop filter whose frequency characteristic is changed by the physical feature amount of 3.
[0021]
Based on the first coefficient seed data stored in the storage means and the second to Nth coefficient seed data, the values of the first parameter and the second to Nth parameters set by the parameter setting means are set. Corresponding estimation formula coefficient data is generated.
[0022]
In addition, a plurality of first information data located around the point of interest related to the second information signal are selected from the first information signal. Then, from the plurality of pieces of first information data and coefficient data of the estimation formula, the information data of the attention point related to the second information signal is calculated using the estimation formula.
[0023]
In this way, a drop filter in which the coefficient data of the estimation formula used when converting the first information signal into the second information signal is changed in frequency characteristics from the teacher signal according to a plurality of physical feature amounts, respectively. Is generated on the basis of a plurality of types of coefficient seed data generated by learning between a plurality of types of student signals and teacher signals obtained by using, thereby increasing the capacity of a memory for storing coefficient data Therefore, coefficient data corresponding to the change of the parameter value can be obtained, and therefore, the parameter value can be adjusted to arbitrarily adjust the output quality by the second information signal, and various changes can be made. The coefficient data to be obtained can be obtained, and therefore the quality of the output by the second information signal can be variously adjusted.
[0024]
The coefficient seed data generating apparatus according to the present invention obtains coefficient data of an estimation formula used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data. An apparatus for generating coefficient seed data to be used, which is obtained from a teacher signal corresponding to a second information signal by using a drop filter whose frequency characteristic is changed by a first physical feature amount. A first data selection means for selecting a plurality of information data located around the attention point related to the teacher signal from a plurality of student signals corresponding to the information signal; and a plurality of data selected by the first data selection means The learning data is generated based on the information data and the information data of the attention point related to the teacher signal, and includes the first parameter indicating the first physical feature amount based on the generated learning data. From the first coefficient data generating means for generating the first coefficient seed data which is the coefficient data of the generation formula for generating the coefficient data of the estimation formula, and the teacher signal corresponding to the second information signal, Each of the N-th physical features (N is an integer of 2 or more) can be obtained by using a drop filter in which the frequency characteristics are changed from frequency characteristics corresponding to predetermined values of other physical characteristics. The second data selection means for selecting a plurality of information data located around the attention point related to the teacher signal from the plurality of student signals corresponding to the first information signal, and the second data selection means Learning data is generated based on the plurality of pieces of information data and the information data of the attention point related to the teacher signal, and the second to N-th physical feature amounts are generated based on the generated learning data. ~ Nth parameter Hints in which and a second coefficient data generation means for generating coefficient seed data of second to N is a coefficient data production equation for producing coefficient data of said estimated equation.
[0025]
Also, the coefficient seed data generation method according to the present invention calculates coefficient data of an estimation formula used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data. A method of generating coefficient seed data used sometimes, which is obtained by using a drop filter in which a frequency characteristic is changed by a first physical feature amount from a teacher signal corresponding to a second information signal. A first step of selecting a plurality of pieces of information data located around a point of interest related to the teacher signal from a plurality of student signals corresponding to the information signal, a plurality of pieces of information data selected in the first step, and Learning data is generated based on the attention point information data related to the teacher signal, and the estimation formula includes a first parameter indicating a first physical feature amount based on the generated learning data. From the second step of generating the first coefficient seed data which is the coefficient data of the generation formula for generating the coefficient data, and the teacher signal corresponding to the second information signal, the second to Nth (N is 2 The first information signal obtained by using a drop filter in which the frequency characteristic is changed from the frequency characteristic corresponding to a predetermined value of the other physical feature quantity by the physical feature quantity of the above integer) A third step of selecting a plurality of information data located around the attention point related to the teacher signal from a plurality of student signals corresponding to the teacher signal, and a plurality of information data and a teacher signal selected in the third step Learning data is generated based on the information data of the attention point, and the estimation formula includes the second to Nth parameters indicating the second to Nth physical feature amounts based on the generated learning data. Coefficient data A coefficient data generation equation for generating to those and a fourth step of generating coefficient seed data of second to N.
[0027]
In the present invention, a plurality of student signals corresponding to the first information signal using a drop filter whose frequency characteristic is changed by the first physical feature amount from the teacher signal corresponding to the second information signal. Is obtained. A plurality of pieces of information data located around the attention point related to the teacher signal are selected from the student signal. Learning data is generated based on the information data of the plurality of information data and the attention point related to the teacher signal. Based on the learning data, the estimation data including the first parameter indicating the first physical feature amount is included. First coefficient seed data, which is coefficient data of a generation formula for generating coefficient data, is generated.
[0028]
Similarly, from the teacher signal corresponding to the second information signal, this frequency characteristic is obtained from the frequency characteristic corresponding to the predetermined value of the other physical feature quantity by the second to Nth physical feature quantities. A plurality of student signals corresponding to the first information signal are obtained using a drop filter in which is changed. A plurality of pieces of information data located around the attention point related to the teacher signal are selected from the student signal. Learning data is generated based on the plurality of information data and the information data of the attention point related to the teacher signal, and the second to Nth physical feature quantities indicating the second to Nth physical feature amounts are generated based on the generated learning data. Second to Nth coefficient seed data that is coefficient data of a generation formula for generating coefficient data of the estimation formula including the Nth parameter is generated.
[0029]
The first parameter used when converting the first information signal into the second information signal by using the first coefficient seed data and the second to Nth coefficient seed data generated in this way. And coefficient data of the estimation formula corresponding to the values of the second and Nth parameters.
[0030]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a configuration of a television receiver 100 as an embodiment. The television receiver 100 obtains an SD signal called a 525i signal from a broadcast signal, converts the 525i signal into an HD signal called a 1050i signal, and displays an image based on the HD signal.
[0031]
The television receiver 100 includes a microcomputer, and includes a system controller 101 for controlling the operation of the entire system, and a remote control signal receiving circuit 102 that receives a remote control signal. The remote control signal receiving circuit 102 is connected to the system controller 101, receives a remote control signal RM output from the remote control transmitter 200 according to a user operation, and supplies an operation signal corresponding to the signal RM to the system controller 101. Is configured to do.
[0032]
Further, the television receiver 100 is supplied with a receiving antenna 105 and a broadcast signal (RF modulated signal) captured by the receiving antenna 105, and performs a channel selection process, an intermediate frequency amplification process, a detection process, and the like. A tuner 106 for obtaining a signal (525i signal) and a buffer memory 109 for temporarily storing an SD signal output from the tuner 106 are provided.
[0033]
Further, the television receiver 100 converts the SD signal (525i signal) temporarily stored in the buffer memory 109 into an HD signal (1050i signal), and an output from the image signal processing unit 110. A display unit 111 for displaying an image by an HD signal, an OSD (On Screen Display) circuit 112 for generating a display signal SCH for displaying a character figure or the like on the screen of the display unit 111, and And a synthesizer 113 for synthesizing the display signal SCH with the HD signal output from the image signal processing unit 110 and supplying the display signal SCH to the display unit 111. The display unit 111 is configured by a flat panel display such as a CRT (cathode-ray tube) display or an LCD (liquid crystal display).
[0034]
The operation of the television receiver 100 shown in FIG. 1 will be described.
The SD signal (525i signal) output from the tuner 106 is supplied to the buffer memory 109 and temporarily stored. The SD signal temporarily stored in the buffer memory 109 is supplied to the image signal processing unit 110 and converted into an HD signal (1050i signal). That is, the image signal processing unit 110 obtains pixel data (hereinafter referred to as “HD pixel data”) that constitutes an HD signal from pixel data (hereinafter referred to as “SD pixel data”) that constitutes an SD signal. The HD signal output from the image signal processing unit 110 is supplied to the display unit 111, and an image based on the HD signal is displayed on the screen of the display unit 111.
[0035]
The user can change the image quality of the image displayed on the screen of the display unit 111 as described above by operating the remote control transmitter 200. As will be described later, in the image signal processing unit 110, the HD pixel data is calculated by an estimation formula. As the coefficient data of this estimation formula, a value h adjusted by the user's operation of the remote control transmitter 200 is used. ₁ , V ₁ , H ₂ , V ₂ Those corresponding to are generated and used. Where h ₁ , V ₁ Constitutes the value of the first parameter, h ₂ , V ₂ Constitutes the value of the second parameter.
[0036]
Thereby, the image quality of the HD signal output from the image signal processing unit 110 is adjusted to the adjusted value h. ₁ , V ₁ , H ₂ , V ₂ It becomes a thing corresponding to. In the parameter adjustment state, the parameter value is displayed on the screen of the display unit 111. The user can adjust the value of the parameter with reference to this display.
[0037]
Next, details of the image signal processing unit 110 will be described.
The image signal processing unit 110 selectively extracts data of a plurality of SD pixels located around the target pixel related to the HD signal (1050i signal) from the SD signal (525i signal) stored in the buffer memory 109. The first to third tap selection circuits 121 to 123 for outputting are provided.
[0038]
The first tap selection circuit 121 selectively extracts data of SD pixels (referred to as “prediction taps”) used for prediction. The second tap selection circuit 122 selectively extracts data of SD pixels (referred to as “space class taps”) used for class classification corresponding to the level distribution pattern of the SD pixel data. The third tap selection circuit 123 selectively extracts data of SD pixels (referred to as “motion class taps”) used for class classification corresponding to motion. When the space class is determined using SD pixel data belonging to a plurality of fields, motion information is also included in this space class.
[0039]
Further, the image signal processing unit 110 detects the level distribution pattern of the space class tap data (SD pixel data) selectively extracted by the second tap selection circuit 122, and determines the space class based on the level distribution pattern. It has a spatial class detection circuit 124 that detects and outputs the class information.
[0040]
In the space class detection circuit 124, for example, an operation is performed to compress each SD pixel data from 8-bit data to 2-bit data. The space class detection circuit 124 outputs compressed data corresponding to each SD pixel data as class information of the space class. In the present embodiment, data compression is performed by ADRC (Adaptive Dynamic Range Coding). As the information compression means, DPCM (predictive coding), VQ (vector quantization), or the like may be used in addition to ADRC.
[0041]
Originally, ADRC is an adaptive requantization method developed for high performance coding for VTR (Video Tape Recorder), but it can efficiently express local patterns of signal level with short word length. It is suitable for use in data compression. When ADRC is used, the maximum value of space class tap data (SD pixel data) is MAX, the minimum value is MIN, the dynamic range of space class tap data is DR (= MAX−MIN + 1), and the number of requantization bits Is P, the requantized code qi as compressed data is obtained by the calculation of the equation (1) for each SD pixel data ki as the space class tap data. However, in the expression (1), [] means a truncation process. When there are Na SD pixel data as the space class tap data, i = 1 to Na.
qi = [(ki-MIN + 0.5). 2 ^P / DR] (1)
[0042]
The image signal processing unit 110 detects a motion class mainly representing the degree of motion from the motion class tap data (SD pixel data) selectively extracted by the third tap selection circuit 123, and A motion class detection circuit 125 that outputs class information is provided.
[0043]
In the motion class detection circuit 125, the inter-frame difference is calculated from the motion class tap data (SD pixel data) mi, ni selectively extracted by the third tap selection circuit 123, and the average of absolute values of the differences is calculated. Threshold processing is performed on the value to detect a motion class that is an index of motion. That is, in the motion class detection circuit 125, the average value AV of the absolute value of the difference is calculated by the equation (2). For example, when 12 pieces of SD pixel data m1 to m6 and n1 to n6 are extracted by the third tap selection circuit 123 as described above, Nb in the expression (2) is 6.
[0044]
[Expression 1]

[0045]
Then, in the motion class detection circuit 125, the average value AV calculated as described above is compared with one or a plurality of threshold values to obtain class information MV of the motion class. For example, three thresholds th1, th2, th3 (th1 <th2 <th3) are prepared, and when four motion classes are detected, when AV ≦ th1, MV = 0 and th1 <AV ≦ th2 Is MV = 2 when MV = 1, th2 <AV ≦ th3, and MV = 3 when th3 <AV.
[0046]
Also, the image signal processing unit 110 is based on the requantization code qi as the class information of the space class output from the space class detection circuit 124 and the class information MV of the motion class output from the motion class detection circuit 125. A class synthesis circuit 126 is provided for obtaining a class code CL indicating a class to which a pixel (target pixel) of an HD signal (1050i signal) to be created belongs.
In the class synthesis circuit 126, the calculation of the class code CL is performed by the equation (3). In equation (3), Na represents the number of space class tap data (SD pixel data), and P represents the number of requantization bits in ADRC.
[0047]
[Expression 2]

[0048]
The image signal processing unit 110 has a coefficient memory 134. The coefficient memory 134 stores, for each class, a plurality of coefficient data Wi of an estimation formula used in an estimated prediction calculation circuit 127 described later. This coefficient data is information for converting an SD signal (525i signal) into an HD signal (1050i signal). The class code CL output from the above class synthesis circuit 126 is supplied to the coefficient memory 134 as read address information. The coefficient memory 134 provides coefficient data Wi (i = 1 to n) of the estimation formula corresponding to the class code CL. ) Is read out and supplied to the estimated prediction calculation circuit 127.
[0049]
Further, the image signal processing unit 110 has an information memory bank 135. In the estimated prediction calculation circuit 127 described later, the HD pixel data y to be generated is calculated from the prediction tap data (SD pixel data) xi and the coefficient data Wi read from the coefficient memory 134 by the estimation expression (4). Calculated. When the number of predicted taps selected by the first tap selection circuit 121 is 10, n in the equation (4) is 10.
[0050]
[Equation 3]

[0051]
In the information memory bank 135, coefficient seed data w1 as the first coefficient seed data used when generating coefficient data Wi (i = 1 to n) of this estimation formula. _i0 ~ W1 _i9 (I = 1 to n) and coefficient seed data w2a as the second coefficient seed data _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 (I = 1 to n) are stored.
[0052]
As described above, when the 525i signal is converted into the 1050i signal, it is necessary to obtain four pixels of the 1050i signal corresponding to one pixel of the 525i signal in each of the odd and even fields. In this case, the four pixels in the 2 × 2 unit pixel block constituting the 1050i signal in each of the odd and even fields have different phase shifts with respect to the central prediction tap.
[0053]
FIG. 2 shows four-pixel HD in a 2 × 2 unit pixel block constituting a 1050i signal in an odd field. ₁ ~ HD _Four Center prediction tap SD ₀ The phase shift from is shown. Where HD ₁ ~ HD _Four The position of each is SD ₀ K horizontally from the position of ₁ ~ K _Four , M in the vertical direction ₁ ~ M _Four It is only shifted.
[0054]
FIG. 3 shows four-pixel HD in a 2 × 2 unit pixel block constituting a 1050i signal in an even field. ₁ '~ HD _Four Center prediction tap SD in ′ ₀ The phase shift from ′ is shown. Where HD ₁ '~ HD _Four The positions of ′ are respectively SD ₀ Horizontally from the position of ′ ₁ ′ 〜K _Four ′, M in the vertical direction ₁ '~ M _Four It is shifted by ′.
[0055]
Therefore, the information memory bank 135 described above includes a class and an output pixel (HD ₁ ~ HD _Four , HD ₁ '~ HD _Four ′) Coefficient seed data w1 for each combination _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 (I = 1 to n) are stored.
[0056]
FIG. 4 shows coefficient seed data w1 _i0 ~ W1 _i9 Shows the generation method.
First, a plurality of SD signals are generated from the HD signal. In this case, the parameters h and v for changing the horizontal frequency characteristic and the vertical frequency characteristic of the drop filter used when obtaining the SD signal from the HD signal are changed in five stages of 4, 5, 6, 7, and 8, respectively. A total of 25 types of SD signals are generated. Next, learning is performed between the plurality of generated SD signals and HD signals to obtain coefficient seed data w1. _i0 ~ W1 _i9 Is generated.
[0057]
FIG. 5 shows horizontal and vertical frequency characteristics in the drop filter when the parameters h and v are 4, 5, 6, 7, and 8. At this time, every time the values of the parameters h and v change by “1”, the bandwidth as the first physical feature value changes by a certain width, and the bandwidth becomes narrower as the values of the parameters h and v become smaller. . When the sampling frequency of the image is fs and the sampling frequency of the filter is f, the horizontal axis represents 2f / fs. Incidentally, the drop filter when the parameters h and v are 6 has a half band filter because 2f / fs is 0.5 at a gain of 0.5.
[0058]
Since all of the drop filters having the frequency characteristics shown in FIG. 5 are configured to pass up to a certain frequency f, when these filters are used, all signal components in a frequency band lower than the frequency f pass and a frequency band higher than the frequency f is passed. This signal component cannot pass through. Therefore, these filters generate student signals with different degrees of blurring at the edge portions of the image. When learning is performed using these student signals, coefficients for obtaining coefficient data Wi of an estimation expression that emphasizes the edges are obtained. Species data w1 _i0 ~ W1 _i9 Can be obtained.
[0059]
Coefficient seed data w1 _i0 ~ W1 _i9 Explain how to find out.
Here, the coefficient data Wi (i = 1 to n) of the estimation formula of the above-described formula (4) is expressed by an approximation formula of the formula (5). Coefficient data w in equation (5) _i0 ~ W _i9 Ask for.

[0060]
Here, for the following explanation, tj (j = 0 to 9) is defined as in the equation (6).

Using this equation (6), equation (5) can be rewritten as equation (7).
[0061]
[Expression 4]

[0062]
Finally, the undetermined coefficient w is obtained by learning using the 25 types of SD signals (see FIG. 4) generated as described above. _ij Ask for. That is, for each class and output pixel, a coefficient value that minimizes the square error is determined using a plurality of SD pixel data and HD pixel data. This is a so-called least square method. The learning number is m, and the residual in the kth learning data (1 ≦ k ≦ m) is e. _k When E is the sum of square errors, E is expressed by equation (8) using equations (4) and (5). Where x _ik Is the k-th pixel data at the i-th predicted tap position of the SD image, y _k Represents pixel data of the k-th HD image corresponding thereto.
[0063]
[Equation 5]

[0064]
In the least squares method, w in equation (8) is used. _ij W such that the partial differential due to becomes zero _ij Ask for. This is shown by equation (9).
[0065]
[Formula 6]

[0066]
Hereinafter, as in the equations (10) and (11), X _ipjq , Y _ip Is defined, equation (9) can be rewritten as equation (12) using a matrix.
[0067]
[Expression 7]

[0068]
[Equation 8]

[0069]
This equation is generally called a normal equation. This normal equation can be calculated using the sweep-out method (Gauss-Jordan elimination method), etc. _ij Is solved and coefficient data w _i0 ~ W _i9 Is calculated. Coefficient data w calculated in this way _i0 ~ W _i9 Is coefficient seed data w1 _i0 ~ W1 _i9 It becomes.
Coefficient seed data w1 obtained in this way _i0 ~ W1 _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents a base curved surface as shown in FIG.
[0070]
FIG. 6 shows coefficient seed data w2a _i0 ~ W2a _i9 Shows the generation method.
First, a plurality of SD signals are generated from the HD signal. In this case, the parameter h for changing the horizontal frequency characteristic of the drop filter used for obtaining the SD signal from the HD signal is varied in five stages of 4, 5, 6, 7 and 8, and the vertical frequency characteristic of the drop filter is changed. The variable parameter v is varied in six stages of 1, 2, 3, 4, 5, and 6 to generate a total of 30 types of SD signals. Next, learning is performed between the plurality of generated SD signals and HD signals to obtain coefficient seed data w2a. _i0 ~ W2a _i9 Is generated.
[0071]
FIG. 7 shows the vertical frequency characteristics of the drop filter when the parameter v is 1, 2, 3, 4, 5, 6. At this time, every time the value of the parameter v changes by “1”, the shoulder characteristic as the second physical feature quantity changes by a certain width, and the shoulder portion becomes smoother as the value of the parameter v becomes smaller. When the sampling frequency of the image is fs and the sampling frequency of the filter is f, the horizontal axis represents 2f / fs.
[0072]
Each of the frequency characteristic drop filters shown in FIG. 7 has a configuration in which the low-frequency pass portion is narrowed while allowing the high-frequency pass to some extent, so that when these filters are used, the edge portion remains and the entire image is slightly faded. It becomes out of focus. Accordingly, student signals with different degrees of blurring of the entire image are generated by these filters, and when learning is performed using these student signals, coefficient types for obtaining coefficient data Wi of an estimation expression that emphasizes the entire image. Data w2a _i0 ~ W2a _i9 Can be obtained.
The horizontal frequency characteristic in the drop filter when the parameter h is 4, 5, 6, 7, 8 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is similar to the case of generating (see the broken lines in FIGS. 5 and 7).
[0073]
Here, the vertical frequency characteristic in the drop filter when v = 6 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is equal to the vertical frequency characteristics of the drop filter when v = 6. As a result, the shoulder characteristic (second physical feature value) is changed from the frequency characteristic when the band (first physical feature value) is a predetermined value (v = 6). .
[0074]
Coefficient seed data w2a _i0 ~ W2a _i9 Although the detailed description is omitted, the above-described coefficient seed data w1 is obtained. _i0 ~ W1 _i9 It is the same as the case where it asks for. However, here, the undetermined coefficient w is obtained by learning using the 30 types of SD signals (see FIG. 6) generated as described above. _ij Will be asked.
Coefficient seed data w2a obtained in this way _i0 ~ W2a _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents a derived curved surface 3 as shown in FIG.
[0075]
FIG. 8 shows coefficient seed data w2b _i0 ~ W2b _i9 Shows the generation method.
First, a plurality of SD signals are generated from the HD signal. In this case, the parameter h for changing the horizontal frequency characteristic of the drop filter used when obtaining the SD signal from the HD signal is varied in five stages of 4, 5, 6, 7 and 8, and the vertical frequency characteristic of the drop filter is changed. The variable parameter v is varied in 6 steps of 0, 1, 2, 3, 4, and 5 to generate a total of 30 types of SD signals. Next, learning is performed between the plurality of generated SD signals and HD signals to obtain coefficient seed data w2b. _i0 ~ W2b _i9 Is generated.
[0076]
FIG. 9 shows the vertical frequency characteristics in the drop filter when the parameter v is 0, 1, 2, 3, 4, and 5. At this time, every time the value of the parameter v changes by “1”, the shoulder characteristic as the second physical feature quantity changes by a certain width, and the shoulder portion becomes smoother as the value of the parameter v becomes smaller. When the sampling frequency of the image is fs and the sampling frequency of the filter is f, the horizontal axis represents 2f / fs. The horizontal frequency characteristic in the drop filter when the parameter h is 4, 5, 6, 7, 8 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is similar to the case of generating (see the broken lines in FIGS. 5 and 9).
[0077]
Here, the vertical frequency characteristic in the drop filter when v = 5 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is equal to the vertical frequency characteristic of the drop filter when v = 5. Thus, the shoulder characteristic (second physical feature value) is changed from the frequency characteristic when the band (first physical feature value) is a predetermined value (v = 5). .
[0078]
Coefficient seed data w2b _i0 ~ W2b _i9 Although the detailed description is omitted, the above-described coefficient seed data w1 is obtained. _i0 ~ W1 _i9 It is the same as the case where it asks for. However, here, the undetermined coefficient w is obtained by learning using the 30 types of SD signals (see FIG. 8) generated as described above. _ij Will be asked.
Coefficient seed data w2b calculated in this way _i0 ~ W2b _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents the derived curved surface 2 as shown in FIG.
[0079]
FIG. 10 shows coefficient seed data w2c _i0 ~ W2c _i9 Shows the generation method.
First, a plurality of SD signals are generated from the HD signal. In this case, the parameter h for changing the horizontal frequency characteristic of the drop filter used for obtaining the SD signal from the HD signal is varied in five stages of 4, 5, 6, 7 and 8, and the vertical frequency characteristic of the drop filter is changed. The variable parameter v is varied in six stages of -1, 0, 1, 2, 3, and 4, and a total of 30 types of SD signals are generated. Next, learning is performed between the plurality of generated SD signals and HD signals to obtain coefficient seed data w2c. _i0 ~ W2c _i9 Is generated.
[0080]
FIG. 11 shows the vertical frequency characteristics of the drop filter when the parameter v is −1, 0, 1, 2, 3, 4. At this time, every time the value of the parameter v changes by “1”, the shoulder characteristic as the second physical feature quantity changes by a certain width, and the shoulder portion becomes smoother as the value of the parameter v becomes smaller. When the sampling frequency of the image is fs and the sampling frequency of the filter is f, the horizontal axis represents 2f / fs. The horizontal frequency characteristic in the drop filter when the parameter h is 4, 5, 6, 7, 8 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is similar to the case of generating (see the broken lines in FIGS. 5 and 11).
[0081]
Here, the vertical frequency characteristic in the drop filter when v = 4 is the coefficient seed data w1 described above. _i0 ~ W1 _i9 Is equal to the vertical frequency characteristic of the drop filter when v = 4. As a result, the shoulder characteristic (second physical feature value) is changed from the frequency characteristic when the band (first physical feature value) is a predetermined value (v = 4). .
[0082]
Coefficient seed data w2c _i0 ~ W2c _i9 Although the detailed description is omitted, the above-described coefficient seed data w1 is obtained. _i0 ~ W1 _i9 It is the same as the case where it asks for. However, here, the undetermined coefficient w is obtained by learning using the 30 types of SD signals (see FIG. 10) generated as described above. _ij Will be asked.
Coefficient seed data w2c calculated in this way _i0 ~ W2c _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents the derived curved surface 1 as shown in FIG.
[0083]
Coefficient seed data w2a mentioned above _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 Are obtained from the student signal generated by the filter whose vertical frequency characteristics are changed by the shoulder characteristics (second physical feature amount), and the curved surfaces of the derived curved surfaces 1 to 3 are similar. It becomes.
[0084]
By fixing the parameter h, the coefficient space as shown in FIG. 12 can be represented by a plane as shown in FIG. FIG. 14 shows a correspondence relationship between the band and shoulder characteristics of the drop filter and the coefficient data Wi. The left coefficient data A to I correspond to the band and shoulder characteristics of the right A to I, respectively.
[0085]
Returning to FIG. 1, the image signal processing unit 110 adjusts the value h adjusted for each combination of class and output pixel based on the coefficient seed data stored in the information memory bank 135. ₁ , V ₁ , H ₂ , V ₂ And a coefficient generation circuit 136 for calculating coefficient data Wi of the estimation formula corresponding to the above.
[0086]
Here, calculation of the coefficient data Wi of the estimation formula in the coefficient generation circuit 136 will be described.
FIG. 15 shows the flow of specifying the coefficient data Wi in the coefficient space. First, the value h ₁ , V ₁ To specify a point on the base surface, then the value h ₂ , V ₂ Specifies a point on the derived surface that contains a point on this specified base surface. The point on this specified derived surface is the value h ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi to be obtained corresponding to. As described above, the direction of changing the shoulder characteristics is the vertical direction, and the coefficient seed data w2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 H. ₁ = H ₂ It is.
[0087]
Value h ₁ , V ₁ , H ₂ , V ₂ The method for obtaining the coefficient data Wi corresponding to is described.
First, the value h ₁ , V ₁ A case where the second coefficient seed data corresponding to the derived curved surface (the derived curved surface of interest) including the point on the base curved surface specified by is present in the information memory bank 135 will be described. In this case, the second coefficient seed data is expressed as coefficient data w in equation (5). _i0 ~ W _i9 And the parameters h and v in equation (5) ₂ , V ₂ Is assigned to obtain coefficient data Wi.
[0088]
Then the value h ₁ , V ₁ A case where the second coefficient seed data corresponding to the derived curved surface (the target derived curved surface) including the point on the base curved surface specified by the information memory bank 135 does not exist will be described. In this case, coefficient data Wi is obtained by interpolation processing.
[0089]
As shown in FIG. 16, the value h ₁ , V ₁ The interpolation process will be described with reference to FIG. 17, taking as an example the case where the derived curved surface (the target derived curved surface) including the point on the base curved surface specified by is located between the derived curved surfaces 2 and 3.
In this case, the value h ₁ , V ₁ , H ₂ , V ₂ If the coefficient data Wi corresponding to is X, X is calculated by equation (13).
X = (Wa * b + Wb * a) / (a + b) (13)
[0090]
Here, Wa is coefficient seed data w2a corresponding to the derived curved surface 3. _i0 ~ W2a _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 And the parameters h and v in equation (5) ₂ , Va is substituted to obtain the coefficient data Wi. However, va = v ₂ + (6-v ₁ ).
[0091]
Wb is coefficient seed data w2b corresponding to the derived curved surface 2. _i0 ~ W2b _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 And the parameters h and v in equation (5) ₂ , Vb is substituted coefficient data Wi obtained. However, vb = v ₂ − (V ₁ -5).
[0092]
The coefficient seed data w1 corresponding to the base curved surface _i0 ~ W1 _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 Is applied to the parameters h and v in the equation (5). ₁ , 5 "," h ₁ , V ₁ "," H ₁ , 6 "is substituted for coefficient data Wi _Five , W _v1 , W ₆ Where a = W ₆ -W _v1 , B = W _v1 -W _Five It is.
[0093]
The value h ₁ , V ₁ When the derived curved surface (the target derived curved surface) including the point on the base curved surface specified by is located between the derived curved surfaces 1 and 2, the value h is obtained by the same interpolation process. ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi corresponding to can be obtained. However, when the target derived curved surface is located outside the derived curved surface 1 or the derived curved surface 3, this interpolation process cannot be used. Therefore, in this embodiment, for example, the value v is such that the target derived curved surface is located between the derived curved surface 1 to the derived curved surface 3. ₁ However, if the target derived curved surface is located outside the derived curved surface 1 or the derived curved surface 3, a parallel movement process described later may be applied.
[0094]
Returning to FIG. 1, the generation of coefficient data Wi of the estimation formula by the coefficient generation circuit 136 is performed, for example, in each vertical blanking period. Thus, the value h is determined by the user's operation of the remote control transmitter 200. ₁ , V ₁ , H ₂ , V ₂ Is changed, the coefficient data Wi of each class stored in the coefficient memory 134 can be immediately changed to the one corresponding to the value, and the image quality adjustment by the user is performed smoothly.
[0095]
In addition, the image signal processing unit 110 includes a normalized coefficient calculation unit 137 that calculates the normalization coefficient S corresponding to the coefficient data Wi (i = 1 to n) obtained by the coefficient generation circuit 136 according to the equation (14). And a normalization coefficient memory 138 for storing the normalization coefficient S. The normalization coefficient memory 138 is supplied with the class code CL output from the above class synthesis circuit 126 as read address information, and the normalization coefficient S corresponding to the class code CL is read from the normalization coefficient memory 138. This is supplied to the normalization operation circuit 128 described later.
[0096]
[Equation 9]

[0097]
Further, the image signal processing unit 110 uses the prediction tap data (SD pixel data) xi selectively extracted by the first tap selection circuit 121 and the coefficient data Wi read from the coefficient memory 134, from the equation (4). The estimation prediction calculation circuit 127 that calculates the data of the pixel (target pixel) of the HD signal to be created is calculated using the estimation formula (1).
[0098]
As described above, when an SD signal (525i signal) is converted to an HD signal (1050i signal), four pixels of the HD signal (HD in FIG. ₁ ~ HD _Four , HD in FIG. ₁ '~ HD _Four Therefore, the estimated prediction calculation circuit 127 generates pixel data for each 2 × 2 unit pixel block constituting the HD signal. That is, in the estimated prediction calculation circuit 127, the first tap selection circuit 121 configures the prediction pixel data xi corresponding to the four pixels (target pixel) in the unit pixel block and the unit memory block from the coefficient memory 134. The coefficient data Wi corresponding to the four pixels to be supplied is supplied, and the data y of the four pixels constituting the unit pixel block ₁ ~ Y _Four Are individually calculated by the estimation equation (4) described above.
[0099]
The image signal processing unit 110 also outputs 4-pixel data y sequentially output from the estimated prediction calculation circuit 127. ₁ ~ Y _Four Is normalized by dividing by the normalization coefficient S corresponding to the coefficient data Wi (i = 1 to n) used for each calculation read out from the normalization coefficient memory 138 ing. Although not described above, the coefficient data Wi of the estimation formula obtained by the coefficient generation circuit 136 is obtained. The obtained coefficient data includes a rounding error, and the total sum of the coefficient data Wi (i = 1 to n) is 1.0. Is not guaranteed to be. Therefore, the data y of each pixel calculated by the estimated prediction calculation circuit 127 ₁ ~ Y _Four Becomes a level fluctuation due to a rounding error. As described above, normalization by the normalization arithmetic circuit 128 can eliminate the fluctuation.
[0100]
The image signal processing unit 110 also normalizes the data y of four pixels in the unit pixel block that is normalized by the normalization calculation circuit 128 and sequentially supplied. ₁ ′ 〜Y _Four ′ Is line-sequentially processed and output in the format of a 1050i signal.
[0101]
Next, the operation of the image signal processing unit 110 will be described.
From the SD signal (525i signal) stored in the buffer memory 109, the second tap selection circuit 122 uses the second tap selection circuit 122 to surround the four pixels (target pixel) in the unit pixel block constituting the HD signal (1050i signal) to be created The data of the space class tap located at (SD pixel data) is selectively extracted. The space class tap data (SD pixel data) selectively extracted by the second tap selection circuit 122 is supplied to the space class detection circuit 124. In this space class detection circuit 124, each of the SD pixel data as space class tap data is subjected to ADRC processing and re-used as class information of the space class (mainly class classification for waveform expression in space). A quantization code qi is obtained (see equation (1)).
[0102]
In addition, from the SD signal (525i signal) stored in the buffer memory 109, the third tap selection circuit 123 uses the four pixels (target pixel) in the unit pixel block constituting the HD signal (1050i signal) to be created. The data (SD pixel data) of the motion class tap located in the vicinity of is selectively extracted. The motion class tap data (SD pixel data) selectively extracted by the third tap selection circuit 123 is supplied to the motion class detection circuit 125. In this motion class detection circuit 125, class information MV of a motion class (mainly class classification for representing the degree of motion) is obtained from each SD pixel data as motion class tap data.
[0103]
The motion information MV and the above-described requantization code qi are supplied to the class synthesis circuit 126. In this class synthesizing circuit 126, from these motion information MV and requantization code qi, for each unit pixel block constituting the HD signal (1050i signal) to be created, four pixels (target pixel) in the unit pixel block are obtained. A class code CL indicating the class to which it belongs is obtained (see equation (3)). The class code CL is supplied to the coefficient memory 134 and the normalized coefficient memory 138 as read address information.
[0104]
The coefficient memory 134 stores, for example, a value h adjusted by the user by the coefficient generation circuit 136 in each vertical blanking period. ₁ , V ₁ , H ₂ , V ₂ Class and output pixel (HD ₁ ~ HD _Four , HD ₁ '~ HD _Four For each combination of ′), coefficient data Wi (i = 1 to n) of the estimation formula is obtained and stored. In this case, the coefficient generation circuit 136 has coefficient seed data w1 as the first coefficient seed data stored in the information memory bank 135. _i0 ~ W1 _i9 (I = 1 to n) and coefficient seed data w2a as the second coefficient seed data _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 Based on (i = 1 to n), the coefficient data Wi of the estimation formula is obtained by interpolation processing or the like as described above.
In the normalized coefficient memory 138, the normalized coefficient S corresponding to the coefficient data Wi (i = 1 to n) of the estimation formula obtained by the coefficient generation circuit 136 as described above is stored in the normalized coefficient calculation unit 137. Generated and stored.
[0105]
As described above, the class code CL is supplied to the coefficient memory 134 as read address information, so that four output pixels corresponding to the class code CL are output from the coefficient memory 134 (HD in the odd field). ₁ ~ HD _Four , Even field is HD ₁ '~ HD _Four The coefficient data Wi of the estimation equation for ′) is read and supplied to the estimated prediction calculation circuit 127. Further, the four pixels (target pixel) in the unit pixel block constituting the HD signal (1050i signal) to be created by the first tap selection circuit 121 from the SD signal (525i signal) stored in the buffer memory 109. Prediction tap data (SD pixel data) located in the vicinity of the area is selectively extracted.
[0106]
In the estimated prediction calculation circuit 127, the prediction tap data (SD pixel data) xi and the coefficient data Wi for four output pixels read from the coefficient memory 134, 4 in the unit pixel block constituting the HD signal to be created. Pixel (attention pixel) data y ₁ ~ Y _Four Is calculated (see equation (4)). Then, the data y of the four pixels in the unit pixel block constituting the HD signal sequentially output from the estimated prediction calculation circuit 127. ₁ ~ Y _Four Is supplied to the normalization operation circuit 128.
[0107]
As described above, the class code CL is supplied to the normalization coefficient memory 138 as read address information, and is output from the normalization coefficient S corresponding to the class code CL, that is, the estimated prediction calculation circuit 127. HD pixel data y ₁ ~ Y _Four The normalization coefficient S corresponding to the coefficient data Wi used for the calculation is read out and supplied to the normalization calculation circuit 128. In the normalization operation circuit 128, the HD pixel data y output from the estimated prediction operation circuit 127 is displayed. ₁ ~ Y _Four Are each divided by the corresponding normalization factor S and normalized. Thereby, the data y due to the rounding error when the coefficient data Wi is obtained by the coefficient generation circuit 136. ₁ ~ Y _Four Level fluctuations are eliminated.
[0108]
In this way, the data y of the four pixels in the unit pixel block which are normalized by the normalization operation circuit 128 and sequentially output. ₁ ′ 〜Y _Four 'Is supplied to the post-processing circuit 129. In the post-processing circuit 129, data y of four pixels in the unit pixel block sequentially supplied from the normalization arithmetic circuit 128 is obtained. ₁ ′ 〜Y _Four 'Is line-sequentialized and output in the format of a 1050i signal. That is, the post-processing circuit 129 outputs a 1050i signal as an HD signal.
[0109]
As described above, in the television receiver 100 shown in FIG. 1, the value h is set for each combination of class and output pixel based on the coefficient seed data stored in the information memory bank 135. ₁ , V ₁ , H ₂ , V ₂ Is obtained by calculation. Coefficient data Wi of the estimation formula for each combination of class and output pixel is stored in the coefficient memory 134. Then, the HD pixel data y is calculated by the estimated prediction calculation circuit 127 using the coefficient data Wi read from the coefficient memory 134 corresponding to the class code CL.
[0110]
Thus, the user has the value h ₁ , V ₁ , H ₂ , V ₂ By adjusting, the image quality of the image obtained by the HD signal can be arbitrarily adjusted. In this case, the value h ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi for each combination of class and output pixel corresponding to is generated and used by the coefficient generation circuit 136 each time, and a memory for storing a large amount of coefficient data is not required.
[0111]
Also, first coefficient seed data generated by learning between a plurality of student signals and the teacher signal obtained by using a drop filter in which the band as the first physical feature amount is changed from the teacher signal. Coefficient seed data as w1 _i0 ~ W1 _i9 And a second coefficient type generated by learning between a plurality of student signals and a teacher signal obtained by using a drop filter in which a shoulder characteristic as a first physical feature amount is changed from the teacher signal Coefficient seed data w2a as data _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 Based on the above, the coefficient data Wi of the estimation formula is generated, and the coefficient data Wi of the estimation formula can be varied in various ways. Therefore, the image quality of the HD signal can be variously adjusted it can.
[0112]
And the value h ₁ , V ₁ If the second coefficient seed data corresponding to the derived curved surface (the derived curved surface of interest) including the point on the base curved surface specified by is not present in the information memory bank 135, the value h ₁ , V ₁ , H ₂ , V ₂ Is obtained by interpolation processing (see equation (13)), the coefficient seed data stored in the information memory bank 135 can be greatly reduced, and the information memory bank 135 Can greatly save memory capacity. For the same reason, the number of learnings associated with the calculation of coefficient seed data can be reduced, leading to improvement in development efficiency.
[0113]
In the above embodiment, the coefficient seed data w2a corresponding to the derived curved surfaces 1 to 3 is stored in the information memory bank 135 as the second coefficient seed data. _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 However, coefficient seed data corresponding to a larger number of derived curved surfaces may be stored and used in the information memory bank 135.
[0114]
In the above embodiment, the coefficient seed data w2a is stored in the information memory bank 135 as the second coefficient seed data. _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 , W2c _i0 ~ W2c _i9 Is stored and the value h ₁ , V ₁ If the second coefficient seed data corresponding to the derived curved surface (the target derived curved surface) including the point on the base curved surface specified by is not present in the information memory bank 135, the value h ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi of the estimation formula corresponding to the above is obtained by interpolation processing. However, for example, the coefficient seed data w2a is stored in the information memory bank 135 as the second coefficient seed data. _i0 ~ W2a _i9 Only the coefficient seed data w2a _i0 ~ W2a _i9 Is the value h ₁ , V ₁ If it does not correspond to a derived surface (point of interest derived surface) including a point on the base surface specified by ₁ , V ₁ , H ₂ , V ₂ It is also conceivable to obtain the coefficient data Wi of the estimation formula corresponding to the above by parallel movement processing.
[0115]
In this case, the information memory bank 135 stores the coefficient seed data w1 as the first coefficient seed data. _i0 ~ W1 _i9 And coefficient seed data w2a as the second coefficient seed data _i0 ~ W2a _i9 Is stored. These coefficient seed data w1 _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 The generation method is the same as described above. Coefficient seed data w1 _i0 ~ W1 _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents a base curved surface as shown in FIG. Coefficient seed data w2a _i0 ~ W2a _i9 Is the coefficient data w of the approximate expression (5). _i0 ~ W _i9 When applied to (5), the equation (5) represents a derived curved surface as shown in FIG.
[0116]
Note that, by fixing the parameter h, the coefficient space as shown in FIG. 18 can be represented by a plane as shown in FIG. FIG. 20 shows a correspondence relationship between the band and shoulder characteristics of the drop filter and the coefficient data Wi. The left coefficient data A to I correspond to the band and shoulder characteristics of the right A to I, respectively.
Thus, the coefficient seed data w1 is stored in the information memory bank 135. _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 Is stored in the coefficient generation circuit 136 as follows.
[0117]
FIG. 21 shows a flow of specifying the coefficient data Wi in the coefficient space. First, the value h ₁ , V ₁ To specify a point on the base surface, then the value h ₂ , V ₂ Specifies a point on the derived surface that contains a point on this specified base surface. The point on this specified derived surface is the value h ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi to be obtained corresponding to. As described above, the direction of changing the shoulder characteristics is the vertical direction, and the coefficient seed data w2a _i0 ~ W2a _i9 H. ₁ = H ₂ It is.
[0118]
Value h ₁ , V ₁ , H ₂ , V ₂ The method for obtaining the coefficient data Wi corresponding to is described.
First, coefficient seed data w2a existing in the information memory bank 135 _i0 ~ W2a _i9 Is the value h ₁ , V ₁ A case in which a derived curved surface including a point on the base curved surface specified by (a focused curved surface of interest) will be described. In this case, coefficient seed data w2a _i0 ~ W2a _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 And the parameters h and v in equation (5) ₂ , V ₂ Is assigned to obtain coefficient data Wi.
[0119]
Next, coefficient seed data w2a existing in the information memory bank 135 _i0 ~ W2a _i9 Is the value h ₁ , V ₁ A case where the derived curved surface including the point on the base curved surface specified by (a derived curved surface of interest) is not described. In this case, the coefficient data Wi is obtained by parallel movement processing. The interpolation process will be described with reference to FIG.
In this case, the value h ₁ , V ₁ , H ₂ , V ₂ If the coefficient data Wi corresponding to is X, X is calculated by equation (15).
X = Wa + (W _v1 -W ₆ (15)
[0120]
Here, Wa is coefficient seed data w2a corresponding to the derived curved surface. _i0 ~ W2a _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 And the parameters h and v in equation (5) ₂ , Va is substituted to obtain the coefficient data Wi. However, va = v ₂ + (6-v ₁ ).
W _v1 , W ₆ Is the coefficient seed data w1 corresponding to the base curved surface _i0 ~ W1 _i9 Is the coefficient data w of equation (5) _i0 ~ W _i9 Is applied to the parameters h and v in the equation (5). ₁ , V ₁ "," H ₁ , 6 ”is obtained as the coefficient data Wi.
[0121]
Thus, the coefficient seed data w1 is stored in the information memory bank 135. _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 Even if is stored, the value h ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi of the estimation formula corresponding to can be obtained satisfactorily. Further, in the case of obtaining the coefficient data Wi of the estimation formula by the parallel movement process in this way, the coefficient type stored in the information memory bank 135 is larger than that of obtaining the coefficient data Wi of the estimation formula by the interpolation process described above. Data can be reduced and memory capacity can be further saved.
In the above-described embodiment, the direction in which the shoulder characteristics are changed is the vertical direction, but the same can be considered even in the horizontal direction.
[0122]
FIG. 23 shows a configuration of a coefficient seed data generation device 150 that generates coefficient seed data stored in the information memory bank 135 of the television receiver 100 described above.
The coefficient data generation apparatus 150 performs an horizontal thinning process and a vertical thinning process on an input terminal 151 to which an HD signal (1050i signal) as a teacher signal is input, and an SD signal as a student signal. And an SD signal generation circuit 152 to be obtained. The SD signal generation circuit 152 is supplied with parameters h and v. Although not shown, the SD signal generation circuit 152 includes horizontal and vertical thinning filters.
[0123]
The horizontal thinning filter reduces the number of pixels in the horizontal direction from the number corresponding to the HD signal to the number corresponding to the SD signal by thinning processing. This horizontal thinning filter also functions as a drop filter, and the frequency characteristic in the horizontal direction changes depending on the supplied parameter h. That is, coefficient seed data w1 _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 When the value of the parameter h is 4, 5, 6, 7, 8, the band changes as shown in FIG.
[0124]
The vertical thinning filter reduces the number of pixels in the vertical direction from the number corresponding to the HD signal to the number corresponding to the SD signal by thinning processing. This vertical thinning filter also functions as a drop filter, and the frequency characteristics in the vertical direction change depending on the value of the supplied parameter v.
[0125]
That is, coefficient seed data w1 _i0 ~ W1 _i9 When the value of the parameter v is 4, 5, 6, 7 or 8, the band changes as shown in FIG. Coefficient seed data w2a _i0 ~ W2a _i9 When the value of the parameter v is 1, 2, 3, 4, 5, 6, the shoulder characteristic changes as shown in FIG. Coefficient seed data w2b _i0 ~ W2b _i9 When the value of the parameter v is 0, 1, 2, 3, 4, 5, the shoulder characteristic changes as shown in FIG. Furthermore, coefficient seed data w2c _i0 ~ W2c _i9 When the value of the parameter v is −1, 0, 1, 2, 3, 4, the shoulder characteristics change as shown in FIG.
[0126]
Further, the coefficient seed data generation device 150 selects data of a plurality of SD pixels located around the target pixel related to the HD signal (1050i signal) from the SD signal (525i signal) output from the SD signal generation circuit 152. The first to third tap selection circuits 153 to 155 for taking out and outputting the data are provided. These first to third tap selection circuits 153 to 155 are configured similarly to the first to third tap selection circuits 121 to 123 of the image signal processing unit 110 described above.
[0127]
The coefficient seed data generation device 150 detects the level distribution pattern of the space class tap data (SD pixel data) selectively extracted by the second tap selection circuit 154, and the space class based on the level distribution pattern. And a space class detection circuit 157 for outputting the class information. The space class detection circuit 157 is configured in the same manner as the space class detection circuit 124 of the image signal processing unit 110 described above. From this space class detection circuit 157, a requantization code qi for each SD pixel data as space class tap data is output as class information indicating a space class.
[0128]
The coefficient seed data generation device 150 detects a motion class mainly representing the degree of motion from the motion class tap data (SD pixel data) selectively extracted by the third tap selection circuit 155, A motion class detection circuit 158 that outputs the class information MV is provided. The motion class detection circuit 158 is configured in the same manner as the motion class detection circuit 125 of the image signal processing unit 110 described above. In this motion class detection circuit 158, the inter-frame difference is calculated from the motion class tap data (SD pixel data) selectively extracted by the third tap selection circuit 155, and the difference between the absolute values of the absolute values of the difference is calculated. Then, threshold processing is performed to detect a motion class that is an index of motion.
[0129]
Also, the coefficient seed data generation device 150 is based on the requantization code qi as the class information of the space class output from the space class detection circuit 157 and the class information MV of the motion class output from the motion class detection circuit 158. And a class synthesis circuit 159 for obtaining a class code CL indicating a class to which the target pixel relating to the HD signal (1050i signal) belongs. The class synthesis circuit 159 is also configured similarly to the class synthesis circuit 126 of the image signal processing unit 110 described above.
[0130]
Further, the coefficient seed data generation device 150 selects each HD pixel data y as target pixel data obtained from the HD signal supplied to the input terminal 151, and a first tap selection corresponding to each HD pixel data y. Prediction tap data (SD pixel data) xi selectively extracted by the circuit 153, the values of parameters h and v, and the class code CL output from the class synthesis circuit 159 corresponding to each HD pixel data y To coefficient class data w for each class _i0 ~ W _i9 A normal equation generation unit 160 that generates a normal equation (see equation (12)) for obtaining (i = 1 to n) is provided.
[0131]
In this case, learning data is generated by a combination of one HD pixel data y and n prediction tap data (SD pixel data) xi corresponding to the HD pixel data y, but this corresponds to changes in the values of the parameters h and v. Thus, the horizontal and vertical frequency characteristics in the SD signal generation circuit 152 are varied, and a plurality of SD signals are sequentially generated. As a result, the normal equation generation unit 160 generates a normal equation in which many learning data having different values of the parameters h and v are registered, and the coefficient seed data w1. _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 Can be obtained respectively.
[0132]
Further, in this case, learning data is generated by a combination of one HD pixel data y and n prediction tap data (SD pixel data) xi corresponding to the HD pixel data y. HD in Figure 2 ₁ ~ HD _Four , HD 1 ′ to HD in FIG. _Four Every time (see '), a normal equation is generated. For example, HD ₁ The normal equation corresponding to, the deviation value for the center prediction tap is HD ₁ Is generated from learning data composed of HD pixel data y having the same relationship.
[0133]
Further, the coefficient seed data generation device 150 is supplied with the data of the normal equation generated for each combination of the class and the output pixel by the normal equation generation unit 160, solves the normal equation, and outputs the combination of the class and the output pixel. And coefficient seed data w1 _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 The coefficient seed data determining unit 161 for obtaining the coefficient seed data, and the coefficient seed memory 162 for storing the obtained coefficient seed data. In the coefficient seed data determining unit 161, the normal equation is solved by, for example, a sweeping out method or the like to obtain coefficient seed data.
[0134]
The operation of the coefficient seed data generation device 150 shown in FIG. 23 will be described.
An HD signal (1050i signal) as a teacher signal is supplied to the input terminal 151, and the HD signal is subjected to horizontal and vertical thinning processing by the SD signal generation circuit 152, and an SD signal (525i) as a student signal. Signal) is generated.
[0135]
From this SD signal (525i signal), the second tap selection circuit 154 selectively extracts data of the space class tap (SD pixel data) located around the target pixel related to the HD signal (1050i signal). The space class tap data (SD pixel data) selectively extracted by the second tap selection circuit 154 is supplied to the space class detection circuit 157. In this space class detection circuit 157, each SD pixel data as space class tap data is subjected to ADRC processing to be re-used as class information of a space class (mainly class classification for waveform expression in space). A quantization code qi is obtained (see equation (1)).
[0136]
Further, from the SD signal generated by the SD signal generation circuit 152, the third tap selection circuit 155 selectively extracts data of the motion class tap (SD pixel data) located around the target pixel related to the HD signal. It is. The motion class tap data (SD pixel data) selectively extracted by the third tap selection circuit 155 is supplied to the motion class detection circuit 158. In this motion class detection circuit 158, class information MV of a motion class (mainly class classification for representing the degree of motion) is obtained from each SD pixel data as motion class tap data.
[0137]
The class information MV and the above-described requantization code qi are supplied to the class synthesis circuit 159. The class synthesis circuit 159 obtains a class code CL indicating the class to which the pixel of interest related to the HD signal (1050i signal) belongs from the class information MV and the requantization code qi (see equation (3)).
[0138]
Further, from the SD signal generated by the SD signal generation circuit 152, the first tap selection circuit 153 selectively extracts data of prediction taps (SD pixel data) located around the target pixel related to the HD signal. .
[0139]
Then, each HD pixel data y as target pixel data obtained from the HD signal supplied to the input terminal 151 and the first tap selection circuit 153 are selectively extracted corresponding to each HD pixel data y. From the prediction tap data (SD pixel data) xi, the values of parameters h and v, and the class code CL output from the class synthesis circuit 159 corresponding to each HD pixel data y, the normal equation generation unit 160 , Coefficient seed data w for each combination of class and output pixel _i0 ~ W _i9 A normal equation (see equation (12)) is obtained individually.
[0140]
Here, coefficient seed data w1 _i0 ~ W1 _i9 When the parameters h and v are generated, the parameters h and v are sequentially changed to 4, 5, 6, 7 and 8, and the vertical and horizontal decimation filter bands are changed as shown in FIG. The normal equation generated by the normal equation generator 160 is the coefficient seed data w1. _i0 ~ W1 _i9 Will be to get.
[0141]
Coefficient seed data w2a _i0 ~ W2a _i9 When the parameter h is generated, the parameter h is sequentially changed as 4, 5, 6, 7, and 8, and the band of the horizontal thinning filter is changed as shown in FIG. , 2, 3, 4, 5 and 6 and the shoulder characteristics of the vertical thinning filter are changed as shown in FIG. Therefore, the normal equation generated by the normal equation generator 160 is the coefficient seed data w2a. _i0 ~ W2a _i9 Will be to get.
[0142]
Coefficient seed data w2b _i0 ~ W2b _i9 When the parameter h is generated, the parameter h is sequentially changed to 4, 5, 6, 7, and 8, and the band of the horizontal thinning filter is changed as shown in FIG. , 1, 2, 3, 4 and 5, and the shoulder characteristics of the vertical thinning filter are changed as shown in FIG. Therefore, the normal equation generated by the normal equation generation unit 160 is the coefficient seed data w2b. _i0 ~ W2b _i9 Will be to get.
[0143]
Coefficient seed data w2c _i0 ~ W2c _i9 When the parameter h is generated, the parameter h is sequentially changed to 4, 5, 6, 7, and 8, and the band of the horizontal thinning filter is changed as shown in FIG. 1, 0, 1, 2, 3, and 4 are sequentially changed, and the shoulder characteristics of the vertical thinning filter are changed as shown in FIG. Therefore, the normal equation generated by the normal equation generating unit 160 is the coefficient seed data w2c. _i0 ~ W2c _i9 Will be to get.
[0144]
Then, the coefficient data determination unit 161 solves each normal equation, and for each combination of class and output pixel, coefficient seed data w1. _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 The coefficient seed data is stored in the coefficient seed memory 162.
[0145]
As described above, in the coefficient seed data generation device 150 shown in FIG. 23, the class and output pixels (HD) stored in the information memory bank 135 of the image signal processing unit 110 in FIG. ₁ ~ HD _Four , HD ₁ '~ HD _Four ′) Coefficient seed data w1 for each combination _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 Can be generated.
Note that the processing in the image signal processing unit 110 in FIG. 1 can be realized by software, for example, by an image signal processing device 300 as shown in FIG.
[0146]
First, the image signal processing apparatus 300 shown in FIG. 24 will be described. The image signal processing apparatus 300 includes a CPU 301 that controls the operation of the entire apparatus, a processing program for the CPU 301, and coefficient seed data used for generating coefficient data Wi of the estimation formula (the image signal processing unit in FIG. 1). 110 (information memory bank 135, the same as the information stored in advance) and the like stored in ROM (read only memory) 302 and RAM (random access memory) 303 constituting the work area of CPU 301. . These CPU 301, ROM 302, and RAM 303 are each connected to a bus 304.
[0147]
The image signal processing apparatus 300 also includes a hard disk drive (HDD) 305 as an external storage device and a floppy (R) disk drive (FDD) 307 that drives a floppy (R) disk 306. These drives 305 and 307 are each connected to a bus 304.
[0148]
In addition, the image signal processing apparatus 300 includes a communication unit 308 that is connected to a communication network 400 such as the Internet by wire or wirelessly. The communication unit 308 is connected to the bus 304 via the interface 309.
[0149]
In addition, the image signal processing device 300 includes a user interface unit. The user interface unit includes a remote control signal receiving circuit 310 that receives a remote control signal RM from the remote control transmitter 200, and a display 311 that includes an LCD (liquid crystal display) or the like. The receiving circuit 310 is connected to the bus 304 via the interface 312, and similarly the display 311 is connected to the bus 304 via the interface 313.
[0150]
Further, the image signal processing apparatus 300 includes an input terminal 314 for inputting an SD signal and an output terminal 315 for outputting an HD signal. The input terminal 314 is connected to the bus 304 via the interface 316, and similarly, the output terminal 315 is connected to the bus 304 via the interface 317.
[0151]
Here, instead of storing the processing program or the like in advance in the ROM 302 as described above, it can also be downloaded from the communication network 400 such as the Internet via the communication unit 308 and stored in the hard disk or RAM 303 for use. . These processing programs, coefficient seed data, and the like may be provided on the floppy (R) disk 306.
[0152]
Further, instead of inputting the SD signal to be processed from the input terminal 314, it may be recorded in advance on a hard disk or downloaded from the communication network 400 such as the Internet via the communication unit 308. Also, instead of outputting the processed HD signal to the output terminal 315 or in parallel therewith, it is supplied to the display 311 to display an image, further stored in a hard disk, or via the communication unit 308 such as the Internet. It may be sent to the communication network 400.
[0153]
With reference to the flowchart of FIG. 25, a processing procedure for obtaining an HD signal from an SD signal in the image signal processing apparatus 300 shown in FIG. 24 will be described.
First, in step ST1, processing is started, and in step ST2, SD pixel data is input in frame units or field units. When the SD pixel data is input from the input terminal 314, the SD pixel data is temporarily stored in the RAM 303. When this SD pixel data is recorded on the hard disk, this SD pixel data is read by the hard disk drive 307 and temporarily stored in the RAM 303. In step ST3, it is determined whether or not processing of all frames or fields of input SD pixel data has been completed. When the process is finished, the process ends in step ST4. On the other hand, when the process is not finished, the process proceeds to step ST5.
[0154]
In step ST5, the image quality designation value input by the user operating the remote control transmitter 200, that is, h ₁ , V ₁ , H ₂ , V ₂ Is read from the RAM 303, for example. In step ST6, coefficient seed data w1 used for generating coefficient data Wi of the estimation formula from, for example, the ROM 302. _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 Is read.
[0155]
Next, in step ST7, the value h read in step ST5 ₁ , V ₁ , H ₂ , V ₂ And the coefficient h for each combination of class and output pixel from the coefficient seed data acquired in step ST6. ₁ , V ₁ , H ₂ , V ₂ The coefficient data Wi of the estimation formula corresponding to is generated.
[0156]
Next, in step ST8, the pixel data of the class tap and the prediction tap are acquired from the SD pixel data input in step ST2 corresponding to each HD pixel data to be generated. In step ST9, it is determined whether or not the processing for obtaining HD pixel data has been completed in all areas of the input SD pixel data. If completed, the process returns to step ST2 and proceeds to the input process of SD pixel data of the next frame or field. On the other hand, when the process has not ended, the process proceeds to step ST10.
[0157]
In step ST10, a class code CL is generated from the SD pixel data of the class tap acquired in step ST9. In step ST11, using the coefficient data corresponding to the class code CL and the SD pixel data of the prediction tap, HD pixel data is generated by the estimation formula, and then the process returns to step ST8 and is the same as described above. Repeat the process.
[0158]
In this way, by performing processing according to the flowchart shown in FIG. 25, it is possible to process the SD pixel data constituting the input SD signal and obtain HD pixel data constituting the HD signal. As described above, the HD signal obtained by such processing is output to the output terminal 315, supplied to the display 311 to display an image, and further supplied to the hard disk drive 305 to be supplied to the hard disk. It is recorded.
Although illustration of the processing device is omitted, the processing in the coefficient seed data generation device 150 in FIG. 23 can be realized by software.
[0159]
Referring to the flowchart of FIG. 26, coefficient seed data w1 used to generate coefficient data Wi of the estimation formula _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 A processing procedure for generating the data will be described.
First, in step ST21, the process is started, and in step ST22, horizontal and vertical frequency characteristics of a drop filter used when generating SD pixel data used for learning are selected. In step ST23, it is determined whether or not learning has been completed for all horizontal and vertical frequency characteristics. If learning has not been completed for all frequency characteristics, the process proceeds to step ST24.
[0160]
In this step ST24, known HD pixel data is input in frame units or field units. In step ST25, it is determined whether or not processing has been completed for all HD pixel data. When the process is completed, the process returns to step ST22, the next frequency characteristic is selected, and the same process as described above is repeated. On the other hand, when not completed, the process proceeds to step ST26.
[0161]
In this step ST26, SD pixel data is generated from the HD pixel data input in step ST24 by using the horizontal and vertical frequency characteristic drop filter selected in step ST22. In step ST27, class tap and prediction tap pixel data corresponding to each HD pixel data input in step ST24 is acquired from the SD pixel data generated in step ST26.
[0162]
In step ST28, it is determined whether or not the learning process has been completed in all regions of the generated SD pixel data. When the learning process is completed, the process returns to step ST24, the next HD pixel data is input, and the same process as described above is repeated. On the other hand, when the learning process is not completed, the process proceeds to step ST29. Proceed to
[0163]
In step ST29, a class code CL is generated from the SD pixel data of the class tap acquired in step ST27. In step ST30, coefficient seed data w _i0 ~ W _i9 To generate a normal equation (see equation (12)). Here, a normal equation is generated individually for each combination of class and output pixel. Thereafter, the process returns to step ST27.
[0164]
Further, when learning is completed for all frequency characteristics in step ST23, the process proceeds to step ST31. In this step ST31, the coefficient seed data w for each combination of class and output pixel is obtained by solving each normal equation by a sweeping method or the like. _i0 ~ W _i9 And the coefficient seed data w is calculated in step ST32. _i0 ~ W _i9 Is stored in memory.
[0165]
Here, the coefficient seed w1 in step ST22. _i0 ~ W1 _i9 By selecting a frequency characteristic for obtaining the coefficient seed data w in step ST30 _i0 ~ W _i9 As coefficient seed w1 _i0 ~ W1 _i9 Is generated and a coefficient seed data w calculated in step ST31 is generated. _i0 ~ W _i9 Is w1 _i0 ~ W1 _i9 It becomes. Similarly, the coefficient type data w calculated in step ST31 is changed by changing the frequency characteristic selected in step ST22. _i0 ~ W _i9 Is w2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 Or w2c _i0 ~ W2c _i9 It becomes.
[0166]
Next, in step ST33, coefficient seed data w1 _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 It is determined whether all of the above are calculated. If not all have been calculated, the process returns to step ST22 and proceeds to a process for calculating the next coefficient seed data. On the other hand, when all are calculated, the process ends in step ST34.
[0167]
In this way, by performing the processing according to the flowchart shown in FIG. 26, it is stored in the information memory bank 135 of the image signal processing unit 110 of FIG. 1 by the same method as the coefficient seed data generation device 150 shown in FIG. Coefficient seed data w1 _i0 ~ W1 _i9 , W2a _i0 ~ W2a _i9 , W2b _i0 ~ W2b _i9 And w2c _i0 ~ W2c _i9 Can be generated.
[0168]
In the above-described embodiment, an example is shown in which two types of band characteristics and shoulder characteristics are used as physical feature amounts for changing the frequency characteristics of the drop filter. However, three or more types of physical feature amounts are used. It can also be used. In this case, a student signal obtained by using a drop filter in which the frequency characteristic is changed from the frequency characteristic corresponding to a predetermined value of each of the other physical feature quantities by the third and subsequent physical feature quantities is used. Thus, the third and subsequent coefficient seed data are calculated.
[0169]
Therefore, the derived curved surface corresponding to the third and subsequent coefficient seed data always intersects the base curved surface and other derived curved surfaces. For example, considering the third coefficient seed data, the derived curved surface corresponding to the third coefficient seed data intersects with the base curved surface corresponding to the first coefficient seed data, as shown in FIG. As shown in 27B, it intersects with the derived curved surface corresponding to the second coefficient seed data.
[0170]
In the case of FIG. 27A, the frequency characteristics of the drop filter when the third physical feature value is a value corresponding to v = 6, the first physical feature value (for example, band) corresponds to v = 6. This is the same as the frequency characteristic of the drop filter when the value is the same. In the case of FIG. 27B, the frequency characteristics of the drop filter when the third physical feature value is a value corresponding to v = 1, the second physical feature value (shoulder characteristic) is v = 1. It is the same as the frequency characteristic of the drop filter when the value corresponds to.
[0171]
The user has an image quality adjustment axis corresponding to the number of physical feature quantities to be used. In this case, as shown in FIG. 27A, the base curved surface (1) intersects with the derived curved surfaces (2), (3), and the base curved surface (1) as shown in FIG. The method of specifying the coefficient data Wi differs when the derived curved surface (2) intersects with the derived curved surface (3).
[0172]
In the case shown in FIG. 27A, as shown in FIG. 28, first, a point on the base curved surface (1) is designated, and a point on the derived curved surface (2) 'or (3)' including the designated point is designated. Is specified. That is, in this case, the adjustment is made by selectively using one of the adjustment axes corresponding to the derived curved surface (2) or (3).
[0173]
In the case shown in FIG. 27B, as shown in FIG. 29, first, a point on the base curved surface (1) is designated, and then a point on the derived curved surface (2) 'including the designated point is designated. Further, a point on the derived curved surface (3) 'including the designated point is designated. That is, in this case, the adjustment is performed using both of the adjustment axes corresponding to the derived curved surfaces (2) and (3).
[0174]
In this way, the image quality can be adjusted with various adjustment axes using the spatial connection of the curved surfaces, and an arbitrary image quality according to the user's preference can be obtained. As described above, even when three or more types of physical feature quantities are used, the designated coefficient data Wi is generated by the above-described interpolation processing, parallel movement processing, and the like, and is used in the estimated prediction calculation circuit 127. be able to.
[0175]
In the above-described embodiment, the information memory bank 135 includes an output pixel (HD ₁ ~ HD _Four , HD ₁ '~ HD _Four ′), Coefficient seed data for generating coefficient data Wi of the estimation formula is stored. By changing the phase of the output pixel with respect to the pixel of the SD signal (525i signal), HD signals other than the 1050i signal can be obtained in the same manner.
Further, in the above-described embodiment, what classifies is shown, but it goes without saying that the present invention can be similarly applied to those not classified.
[0176]
In the above-described embodiment, the linear equation is used as the estimation equation when generating the HD signal. However, the present invention is not limited to this, and for example, a higher-order equation is used as the estimation equation. It may be a thing.
Moreover, although the case where the information signal is an image signal has been shown in the above-described embodiment, the present invention is not limited to this. For example, the present invention can be similarly applied when the information signal is an audio signal.
[0177]
【The invention's effect】
According to this invention, the frequency characteristics of the coefficient data of the estimation formula used when converting the first information signal into the second information signal are changed from the teacher signal according to the plurality of physical feature amounts, respectively. Capacity of memory for storing coefficient data, which is generated based on multiple types of coefficient seed data generated by learning between multiple types of student signals and teacher signals obtained using drop filters Coefficient data corresponding to the change in the parameter value can be obtained without causing an increase. Therefore, the parameter value can be adjusted to arbitrarily adjust the output quality of the second information signal. Therefore, the coefficient data that changes to the second information signal can be adjusted in various ways.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of a television receiver as an embodiment.
FIG. 2 is a diagram illustrating a phase shift (odd field) from a center prediction tap of four pixels in a unit pixel block of an HD signal (1050i signal).
FIG. 3 is a diagram illustrating a phase shift (even field) from a center prediction tap of four pixels in a unit pixel block of an HD signal (1050i signal).
[Figure 4] Coefficient seed data w1 _i0 ~ W1 _i9 It is a figure which shows the production | generation method.
FIG. 5 is a diagram illustrating horizontal and vertical frequency characteristics (band change) in a drop filter when parameters h and v are 4, 5, 6, 7, and 8. FIG.
[Figure 6] Coefficient seed data w2a _i0 ~ W2a _i9 It is a figure which shows the production | generation method.
7 is a diagram showing vertical frequency characteristics (shoulder characteristic change) in a drop filter when a parameter v is 1, 2, 3, 4, 5, 6. FIG.
[Figure 8] Coefficient seed data w2b _i0 ~ W2b _i9 It is a figure which shows the production | generation method.
FIG. 9 is a diagram illustrating vertical frequency characteristics (shoulder characteristic change) in a drop filter when a parameter v is 0, 1, 2, 3, 4, and 5;
[Figure 10] Coefficient seed data w2c _i0 ~ W2c _i9 It is a figure which shows the production | generation method.
FIG. 11 is a diagram illustrating vertical frequency characteristics (shoulder characteristics change) in a drop filter when a parameter v is set to −1, 0, 1, 2, 3, and 4;
FIG. 12 is a schematic diagram of a coefficient space including a base curved surface and a derived curved surface.
FIG. 13 is a schematic diagram of a coefficient space (plane) including a base curved surface and a derived curved surface.
FIG. 14 is a diagram illustrating a correspondence relationship between a drop filter band and shoulder characteristics and coefficient data Wi;
FIG. 15 is a diagram showing a flow of specifying coefficient data Wi in a coefficient space.
FIG. 16 is a diagram illustrating a case where a target derived curved surface exists between derived curved surfaces 2 and 3;
FIG. 17 is a diagram for explaining interpolation processing;
FIG. 18 is a schematic diagram of a coefficient space including a base curved surface and a derived curved surface.
FIG. 19 is a schematic diagram of a coefficient space (plane) including a base curved surface and a derived curved surface.
FIG. 20 is a diagram showing a correspondence relationship between a drop filter band and shoulder characteristics and coefficient data Wi;
FIG. 21 is a diagram illustrating a flow of specifying coefficient data Wi in a coefficient space.
FIG. 22 is a diagram for explaining parallel movement processing;
FIG. 23 is a block diagram illustrating a configuration example of a coefficient seed data generation device.
FIG. 24 is a block diagram illustrating a configuration example of an image signal processing device to be realized by software.
FIG. 25 is a flowchart showing image signal processing;
FIG. 26 is a flowchart showing a coefficient seed data generation process.
FIG. 27 is a schematic diagram of a coefficient space when the third coefficient seed data is considered.
FIG. 28 is a diagram for explaining how to specify coefficient data;
FIG. 29 is a diagram for explaining how to specify coefficient data;
FIG. 30 is a diagram illustrating a pixel position relationship between a 525i signal and a 1050i signal.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 ... Television receiver, 101 ... System controller, 102 ... Remote control signal receiving circuit, 105 ... Receiving antenna, 106 ... Tuner, 110 ... Image signal processing part, 111 ... Display unit 112 ... OSD circuit 121 ... first tap selection circuit 122 ... second tap selection circuit 123 ... third tap selection circuit 124 ... space class detection Circuit: 125... Motion class detection circuit, 126... Class synthesis circuit, 127... Prediction prediction calculation circuit, 128... Normalization calculation circuit, 129. Memory, 135 ... Information memory bank, 136 ... Coefficient generation circuit, 137 ... Normalized coefficient arithmetic unit, 138 ... Normalized coefficient memory, 150 ... Coefficient seed data generation 151, input terminal, 152, SD signal generation circuit, 153, first tap selection circuit, 154, second tap selection circuit, 155, third tap selection circuit. 157... Space class detection circuit, 158... Motion class detection circuit, 159... Class synthesis circuit, 160. Coefficient seed memory, 200... Remote control transmitter, 300... Image signal processing device

Claims (20)

Used when converting a first information signal made up of a plurality of information data into a second information signal made up of a plurality of information data, the information data of the point of interest related to the second information signal is converted into the first information signal A coefficient data generation device that generates coefficient data of an estimation formula for calculating from a plurality of information data extracted from an information signal,
Corresponding to a teacher signal corresponding to the second information signal and the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by a first physical feature amount The coefficient data of the generation formula for generating the coefficient data of the estimation formula including the first parameter indicating the first physical feature amount generated by learning with a plurality of student signals. 1 coefficient seed data, a teacher signal corresponding to the second information signal, and the second to Nth (N is an integer of 2 or more) physical features from the teacher signal. It is generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from a frequency characteristic corresponding to a predetermined value of the physical feature quantity. Respectively, the second to second Storage means for storing second to Nth coefficient seed data that is coefficient data of a generation formula for generating coefficient data of the estimation formula including the second to Nth parameters indicating the physical feature amount ,
Parameter setting means for setting the value of the first parameter and the values of the second to Nth parameters;
Based on the first coefficient seed data and the second to Nth coefficient seed data stored in the storage means, the first parameter and the second to second parameters set by the parameter setting means. A coefficient data generation device comprising: coefficient data generation means for generating coefficient data of the estimation formula corresponding to the value of the N parameter. The storage means stores the second coefficient seed data in addition to the first coefficient seed data.
The parameter setting means sets the value of the first parameter and the value of the second parameter,
The coefficient data generation means includes a value of the first parameter set by the parameter setting means based on the first coefficient seed data and the second coefficient seed data stored in the storage means, and The coefficient data generation apparatus according to claim 1, wherein coefficient data of the estimation formula corresponding to the value of the second parameter is generated. The second coefficient seed data stored in the storage means has a frequency characteristic based on the second physical feature amount from a plurality of frequency characteristics respectively corresponding to a plurality of values of the first parameter. Consists of a plurality of second coefficient seed data related to the student signal generated by the changed drop filter,
The coefficient data generating means is
When the value of the first parameter set by the parameter setting means corresponds to any of the plurality of values, the second physical property is determined from the frequency characteristic corresponding to the value of the first parameter. Based on the second coefficient seed data related to the student signal generated by the drop filter whose frequency characteristic is changed by the feature quantity, the value of the first parameter and the value of the second parameter are calculated by the generation formula. Generate coefficient data of the corresponding estimation formula,
When the value of the first parameter set by the parameter setting means does not correspond to any of the plurality of values, it corresponds to a value before and after the value of the first parameter among the plurality of values. Based on the two second coefficient seed data related to the student signal generated by the drop filter in which the frequency characteristic is changed by the second physical feature amount from the two frequency characteristics, the parameter is calculated by the generation formula. Two coefficient data of the estimation equation corresponding to the value of the second parameter set by the setting means is obtained, and is obtained from the two coefficient data by the generation equation based on the first coefficient seed data. Further, the value of the first parameter and the second value are obtained by interpolation using the coefficient data of the estimation formula corresponding to the value of the first parameter and the previous and subsequent values. Coefficient data generation apparatus according to claim 2, characterized in that to generate the coefficient data of the estimation equation corresponding to the value of the parameter. The second coefficient seed data stored in the storage means has a frequency characteristic changed by the second physical feature amount from a frequency characteristic corresponding to a single value of the first parameter. Consists of a single second coefficient seed data related to the student signal generated by the drop filter,
The coefficient data generating means is
When the value of the first parameter set by the parameter setting means is the single value, based on the single second coefficient seed data, the generating equation Generating coefficient data of the estimation formula corresponding to the value and the value of the second parameter;
When the value of the first parameter set by the parameter setting means is not the single value, the value set by the parameter setting means by the generation formula based on the single second coefficient seed data The coefficient data of the estimation equation corresponding to the value of the second parameter is obtained, and from the coefficient data, the value of the first parameter obtained by the generation equation based on the first coefficient seed data and the value Generating coefficient data of the estimation formula corresponding to the value of the first parameter and the value of the second parameter by a parallel movement process using the coefficient data of the estimation formula corresponding to a single value. The coefficient data generation device according to claim 2, wherein The storage means is
Based on the teacher signal corresponding to the second information signal and the frequency characteristic corresponding to the predetermined value of the first parameter or the predetermined value of the second parameter from the teacher signal, the third physical signal The third physical feature amount generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter whose frequency characteristic is changed according to the feature amount. Third coefficient seed data which is coefficient data of a generation formula for generating coefficient data of the estimation formula including the third parameter shown is further stored,
The parameter setting means further sets the value of the third parameter,
The coefficient data generation means corresponds to the values of the first to third parameters set by the parameter setting means based on the first to third coefficient seed data stored in the storage means. The coefficient data generation apparatus according to claim 2, wherein coefficient data of the estimation formula is generated. The third coefficient seed data stored in the storage means is a drop in which the frequency characteristic is changed by the third physical feature amount from the frequency characteristic corresponding to the predetermined value of the first parameter. It relates to the student signal obtained by the filter,
The coefficient data generation means generates coefficient data of the estimation formula corresponding to the value of the first parameter and the value of the second parameter or the value of the third parameter. 5. The coefficient data generation device according to 5. The third coefficient seed data stored in the storage means is a drop in which the frequency characteristic is changed by the third physical feature amount from the frequency characteristic corresponding to the predetermined value of the second parameter. It relates to the student signal obtained by the filter,
The coefficient data generation device according to claim 5, wherein the coefficient data generation unit generates coefficient data of the estimation formula corresponding to the values of the first to third parameters. Used when converting a first information signal made up of a plurality of information data into a second information signal made up of a plurality of information data, the information data of the point of interest related to the second information signal is converted into the first information signal A coefficient data generation method for generating coefficient data of an estimation formula for calculating from a plurality of information data extracted from an information signal,
Corresponding to a teacher signal corresponding to the second information signal and the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by a first physical feature amount The coefficient data of the generation formula for generating the coefficient data of the estimation formula including the first parameter indicating the first physical feature amount generated by learning with a plurality of student signals. Obtaining the coefficient seed data of 1 from the storage unit;
Based on the teacher signal corresponding to the second information signal and the second to N-th (N is an integer of 2 or more) physical feature amounts, the other physical feature amounts are respectively determined from the teacher signal. Generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from the frequency characteristic corresponding to the value. Second to Nth coefficient seed data, which are coefficient data of a generation formula for generating coefficient data of the estimation formula including the second to Nth parameters indicating the Nth physical feature amount, are stored from the storage unit. A step to obtain,
Setting the value of the first parameter and the values of the second to Nth parameters;
Based on the obtained first coefficient seed data and the second to Nth coefficient seed data, the values corresponding to the set first parameter value and the second to Nth parameter values, A coefficient data generation method comprising: generating coefficient data of an estimation formula. An information signal processing device for converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data,
Corresponding to a teacher signal corresponding to the second information signal and the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by a first physical feature amount The first coefficient data of the generation formula for generating the coefficient data of the estimation formula including the first parameter indicating the first physical feature amount generated by learning with a plurality of student signals. Coefficient type data, a teacher signal corresponding to the second information signal, and the second to N-th (N is an integer of 2 or more) physical feature values from the teacher signal, respectively, Generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from a frequency characteristic corresponding to a predetermined value of a characteristic feature , Respectively, the second to Nth above Storage means for storing coefficient seed data for second to N is a coefficient data production equation for producing coefficient data of said estimated equation includes a parameter of the second to N indicating the physical feature quantity,
Parameter setting means for setting the value of the first parameter and the values of the second to Nth parameters;
Based on the first coefficient seed data and the second to Nth coefficient seed data stored in the storage means, the value of the first parameter set by the parameter setting means and the second Coefficient data generating means for generating coefficient data of the estimation formula corresponding to the value of the Nth parameter;
First data selection means for selecting a plurality of first information data located around a point of interest related to the second information signal from the first information signal;
From the coefficient data generated by the coefficient data generation means and the plurality of first information data selected by the first data selection means, the information data of the attention point is calculated using the estimation formula. Computing means
An information signal processing apparatus comprising: The storage means stores the second coefficient seed data in addition to the first coefficient seed data,
The parameter setting means sets the value of the first parameter and the value of the second parameter,
The coefficient data generation means includes a value of the first parameter set by the parameter setting means based on the first coefficient seed data and the second coefficient seed data stored in the storage means, and Generate coefficient data of the estimation equation corresponding to the value of the second parameter
The information signal processing apparatus according to claim 9. The second coefficient seed data stored in the storage means has a frequency characteristic according to the second physical feature amount from a plurality of frequency characteristics respectively corresponding to a plurality of values of the first parameter. Consists of a plurality of second coefficient seed data related to the student signal generated by the changed drop filter,
The coefficient data generating means is
When the value of the first parameter set by the parameter setting means corresponds to any of the plurality of values, the second physical value is determined from the frequency characteristic corresponding to the value of the first parameter. Based on the second coefficient seed data related to the student signal generated by the drop filter whose frequency characteristic is changed by the feature amount, the value of the first parameter and the value of the second parameter are calculated by the generation formula. Generate coefficient data of the corresponding estimation formula,
When the value of the first parameter set by the parameter setting means does not correspond to any of the plurality of values, it corresponds to a value before and after the value of the first parameter among the plurality of values. Based on the two second coefficient seed data related to the student signal generated by the drop filter in which the frequency characteristic is changed by the second physical feature amount from the two frequency characteristics, the parameter is calculated by the generation formula. Two coefficient data of the estimation equation corresponding to the value of the second parameter set by the setting means is obtained, and is obtained from the two coefficient data by the generation equation based on the first coefficient seed data. Further, the value of the first parameter and the second value are obtained by interpolation using the coefficient data of the estimation formula corresponding to the value of the first parameter and the previous and subsequent values. Generating the coefficient data of the estimation equation corresponding to the value of the parameter
The information signal processing apparatus according to claim 10. The second coefficient seed data stored in the storage means has a frequency characteristic changed by the second physical feature amount from a frequency characteristic corresponding to a single value of the first parameter. It consists of a single second coefficient seed data related to the student signal generated by the drop filter,
The coefficient data generating means is
When the value of the first parameter set by the parameter setting means is the single value, based on the single second coefficient seed data, the generating equation Generating coefficient data of the estimation equation corresponding to the value and the value of the second parameter;
When the value of the first parameter set by the parameter setting means is not the single value, the value set by the parameter setting means by the generation formula based on the single second coefficient seed data The coefficient data of the estimation formula corresponding to the value of the second parameter is obtained, and from the coefficient data, the value of the first parameter obtained from the generation formula based on the first coefficient seed data and the value of the first parameter Coefficient data of the estimation formula corresponding to the value of the first parameter and the value of the second parameter is generated by parallel movement processing using the coefficient data of the estimation formula corresponding to a single value.
The information signal processing apparatus according to claim 10. The storage means is
From the teacher signal corresponding to the second information signal and the teacher signal, a third physical characteristic is obtained from the frequency characteristic corresponding to the predetermined value of the first parameter or the predetermined value of the second parameter. The third physical feature amount generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter whose frequency characteristic is changed according to the amount. Further storing third coefficient seed data that is coefficient data of a generation formula for generating coefficient data of the estimation formula including a third parameter;
The parameter setting means further sets the value of the third parameter,
The coefficient data generation means corresponds to the first to third parameter values set by the parameter setting means based on the first to third coefficient seed data stored in the storage means. Generate coefficient data for the above estimation formula
The information signal processing apparatus according to claim 10. The third coefficient seed data stored in the storage means is a drop in which the frequency characteristic is changed by the third physical feature amount from the frequency characteristic corresponding to the predetermined value of the first parameter. It relates to the student signal obtained by the filter,
The coefficient data generating means generates coefficient data of the estimation equation corresponding to the value of the first parameter and the value of the second parameter or the value of the third parameter.
The information signal processing apparatus according to claim 13. The third coefficient seed data stored in the storage means is a drop in which the frequency characteristic is changed by the third physical feature amount from the frequency characteristic corresponding to the predetermined value of the second parameter. It relates to the student signal obtained by the filter,
The coefficient data generation means generates coefficient data of the estimation formula corresponding to the values of the first to third parameters.
The information signal processing apparatus according to claim 13. A second data selection means for selecting a plurality of second information data located around the point of interest relating to the second information signal from the first information signal; and the second data selection means. Class detecting means for detecting a class to which the attention point belongs based on the plurality of second information data;
The storage means stores the first coefficient seed data and the second to Nth coefficient seed data for each class that can be detected by the class detection means,
The coefficient data generation means includes the estimation equation corresponding to the class detected by the class detection means, the value of the first parameter set by the parameter setting means, and the values of the second to Nth parameters. Generate coefficient data
The information signal processing apparatus according to claim 9. The coefficient data generating means is
For each class, based on the first coefficient seed data and the second to Nth coefficient seed data stored in the storage means, the value of the first parameter set by the parameter setting means and A coefficient data generation unit that generates coefficient data of the estimation formula corresponding to the values of the second to Nth parameters;
A storage unit for storing coefficient data of the estimation formula for each class generated by the coefficient data generation unit;
A coefficient data reading unit that reads out and outputs the coefficient data of the estimation formula corresponding to the class detected by the detection means from the storage unit;
The information signal processing apparatus according to claim 16. An information signal processing method for converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data,
Corresponding to a teacher signal corresponding to the second information signal and the first information signal obtained from the teacher signal using a drop filter whose frequency characteristic is changed by a first physical feature amount The first coefficient data of the generation formula for generating the coefficient data of the estimation formula including the first parameter indicating the first physical feature amount generated by learning with a plurality of student signals. Obtaining the coefficient seed data from the storage unit;
Based on the teacher signal corresponding to the second information signal and the second to N-th (N is an integer of 2 or more) physical feature amounts, the other physical feature amounts are respectively determined from the teacher signal. Generated by learning with a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from the frequency characteristic corresponding to the value. Second to Nth coefficient seed data, which are coefficient data of a generation formula for generating coefficient data of the estimation formula including the second to Nth parameters indicating the Nth physical feature amount, are stored from the storage unit. A step to obtain,
Setting a value of the first parameter and a value of the second to Nth parameters;
Based on the obtained first coefficient seed data and the second to Nth coefficient seed data, the values corresponding to the set first parameter value and the second to Nth parameter values, Generating coefficient data for the estimation equation;
Selecting a plurality of pieces of information data located around a point of interest related to the second information signal from the first information signal;
Calculating the information data of the attention point from the generated coefficient data and the plurality of selected information data using the estimation formula; and
An information signal processing method comprising: An apparatus for generating coefficient seed data used when obtaining coefficient data of an estimation formula used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data. And
From a plurality of student signals corresponding to the first information signal obtained from a teacher signal corresponding to the second information signal using a drop filter whose frequency characteristic is changed by a first physical feature amount, First data selection means for selecting a plurality of pieces of information data located around a point of interest related to the teacher signal;
Learning data is generated based on the plurality of information data selected by the first data selection means and the information data of the attention point related to the teacher signal, and the first physical data is generated based on the generated learning data. First coefficient data generating means for generating first coefficient seed data that is coefficient data of a generation formula for generating coefficient data of the estimation formula including a first parameter indicating a characteristic feature amount;
From the teacher signal corresponding to the second information signal, frequencies corresponding to predetermined values of other physical feature amounts, depending on the second to Nth (N is an integer of 2 or more) physical feature amounts, respectively. A plurality of pieces of information data located around a point of interest related to the teacher signal are selected from a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from the characteristic Second data selection means for
Learning data is generated based on the plurality of information data selected by the second data selection means and the information data of the attention point related to the teacher signal, and the second to second data are generated based on the generated learning data. Second to generate second to Nth coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of the estimation formula, including second to Nth parameters indicating N physical feature amounts Coefficient data generation means
A coefficient seed data generation device comprising: A method for generating coefficient seed data used when calculating coefficient data of an estimation formula used when converting a first information signal composed of a plurality of information data into a second information signal composed of a plurality of information data. ,
From a plurality of student signals corresponding to the first information signal obtained from a teacher signal corresponding to the second information signal using a drop filter whose frequency characteristic is changed by a first physical feature amount, A first step of selecting a plurality of pieces of information data located around a point of interest related to the teacher signal;
Learning data is generated based on the plurality of information data selected in the first step and the information data of the attention point related to the teacher signal, and the first physical data is generated based on the generated learning data. A second step of generating first coefficient seed data which is coefficient data of a generation formula for generating coefficient data of the estimation formula including a first parameter indicating a feature amount;
From the teacher signal corresponding to the second information signal, frequencies corresponding to predetermined values of other physical feature amounts, depending on the second to Nth (N is an integer of 2 or more) physical feature amounts, respectively. A plurality of pieces of information data located around a point of interest related to the teacher signal are selected from a plurality of student signals corresponding to the first information signal obtained by using a drop filter in which the frequency characteristic is changed from the characteristic A third step,
Learning data is generated based on the plurality of information data selected in the third step and the information data of the attention point related to the teacher signal, and based on the generated learning data, the second to Nth Fourth step of generating second to Nth coefficient seed data, which is coefficient data of a generation formula for generating coefficient data of the estimation formula, including second to Nth parameters indicating physical feature amounts When
A coefficient seed data generation method comprising:

Images