JP3777598B2 - Information conversion device and method, coefficient calculation device and method, storage device, recording medium, and program - Google Patents

Description

【０１００】
式（２１）のｅ＾2（ｅの2乗）を最小とする各予測係数ｗiは、ｅ＾2を予測係数ｗi（i=1,2・・・・）で偏微分した式（２３）に変形し、ｉの各値について偏微分値が０となるものとして計算される。
【０１０１】
【数２】

【０１０２】
式（２３）から各予測係数ｗiを定める具体的な手順について説明する。式（２４），式（２５）のようにＸji，Ｙiを定義するとき、式（２３）は、式（２６）の行列式の形に変形される。
【０１０３】
【数３】

【数４】

【数５】

【０１０４】
この式（２６）が、一般に正規方程式と呼ばれるものである。ここで、Ｘji（ｊ，ｉ＝１，２・・・・ｎ），およびＹi （ｉ＝１，２・・・・ｎ）は、教師画像、および、生徒画像に基づいて計算される。すなわち、正規方程式生成部９６は、ＸjiおよびＹiの値を算出して正規方程式からなる式（２６）を決定する。さらに、係数決定部９７は、式（２６）を解いて各予測係数ｗiを決定する。
【０１０５】
次に、図１９のフローチャートを参照して、図１８の学習装置が予測係数を学習する処理について説明する。
【０１０６】
ステップＳ６１において、教師画像としてのコンポーネント信号がNTSCエンコーダ９１、および、正規方程式生成部９６に入力される。ステップＳ６２において、NTSCエンコーダ９１は、入力されたコンポーネント信号よりNTSC方式のコンポジット信号からなる生徒画像を生成して領域抽出部９２，９５に出力する。
【０１０７】
ステップＳ６３において、領域抽出部９２は、領域抽出処理を実行し、クラスタップを抽出してパターン検出部９３に出力する。尚、この処理は、図６のフローチャートのステップＳ４の処理と同じものである。
【０１０８】
ステップＳ６４において、パターン検出部９３は、パターン検出処理を実行し、クラスタップよりクラスコードを決定するためのパターンを検出し、クラスコード決定部９４に出力する。尚、この処理は、図６のフローチャートのステップＳ５の処理と同様であり、図８のフローチャートを参照して説明した処理と同様であるのでその説明は省略する。
【０１０９】
ステップＳ６５において、クラスコード決定部９４は、パターン検出部９３により検出されたパターンに基づいてクラスコード決定処理を実行し、正規方程式生成部９６に出力する。尚、この処理は、図６のフローチャートのステップＳ６の処理と同様であり、図１２のフローチャートを参照して説明した処理と同様であるので、その説明は省略する。
【０１１０】
ステップＳ６６において、領域抽出部９５は、生徒画像の予測タップを抽出して正規方程式生成部９６に出力する。
【０１１１】
ステップＳ６７において、正規方程式生成部９６は、クラスコード決定部９４より入力されたクラスコード、領域抽出部９５より入力された予測タップ、および、教師画像として入力されたコンポーネント信号から上述の式（２６）で示した正規方程式を生成し、クラスコード決定部９４より入力されたクラスコードと共に係数決定部９７に出力する。
【０１１２】
ステップＳ６８において、係数決定部９７は、上述の式（２６）からなる正規方程式を解いて、予測係数を決定し、クラスコードに対応付けてメモリ９８に記憶させる。
【０１１３】
ステップＳ６９において、全ての画素について処理が施されたか否かが判定され、全ての画素について処理が施されていないと判定された場合、その処理は、ステップＳ６３に戻る。すなわち、全ての画素の処理が終了されるまで、ステップＳ６３乃至Ｓ６９の処理が繰り返される。ステップＳ６９において、全ての画素について処理が施されたと判定された場合、その処理は、終了する。
【０１１４】
以上の処理により、着目画素とその近接画素の時間方向、垂直方向、および、水平方向のそれぞれについての簡易Ｙ／Ｃ分離により生成したクロマ信号によりクラスコードを決定することができ、着目画素に対して適切な予測係数を設定することができるので、画素毎に適正なコンポジット信号からコンポーネント信号への変換処理を実行することができ、画質の劣化を抑制することができる。
【０１１５】
また、以上においては、クラスコード決定部５３，９４において、簡易Ｙ／Ｃ分離するときの方向が異なるクロマ信号Ｃの差分の絶対値｜Ｃｖ１−Ｃｔ１｜，｜Ｃｔ１−Ｃｈ１｜，｜Ｃｈ２−Ｃｖ２｜，｜Ｃｖ２−Ｃｔ２｜，｜Ｃｔ２−Ｃｈ２｜，｜Ｃｈ３−Ｃｖ３｜，｜Ｃｖ３−Ｃｔ３｜，｜Ｃｔ３−Ｃｈ３｜毎に所定の閾値ｔｈとの大小の比較により２値のクラスコードを設定する例について説明してきたが、例えば、その差分そのものと、絶対値の等しい２個の閾値ｔｈ，−ｔｈ（｜ｔｈ｜＝｜−ｔｈ｜）との大小関係から３値のクラスコードを決定するようにしても良い。図２０は、異なる方向からＹ／Ｃ分離することにより求められたクロマ信号の差分を求め、２個の閾値ｔｈ，−ｔｈとの大小関係からクラスコードを決定する場合のクラスコード決定部５３の構成を示している。
【０１１６】
異なる方向よりＹ／Ｃ分離されたクロマ信号をＣｍ，Ｃｎ（（Ｃｍ，Ｃｎ）は、（Ｃｖ，Ｃｈ），（Ｃｈ，Ｃｔ）（Ｃｔ，Ｃｖ）のいずれかである）で表すとき、クロマ信号Ｃｍ，Ｃｎが、クラスコード決定部５３に入力されると、減算部１１１は、その差分（Ｃｍ−Ｃｎ）を求めて比較器１１２に出力する。比較器１１２は、入力された差分（Ｃｍ−Ｃｎ）を閾値ｔｈ，−ｔｈと比較する。この場合、比較器１１２は、差分（Ｃｍ−Ｃｎ）が閾値−ｔｈ以下であった場合、クラスコードを０に、差分（Ｃｍ−Ｃｎ）が閾値−ｔｈよりも大きく、閾値ｔｈ以下であった場合、クラスコードを１に、差分（Ｃｍ−Ｃｎ）が閾値ｔｈより大きい場合、クラスコードを２に設定する。このようなクラスコードの設定により、３＾９（３の９乗）個のクラスコードを設定させることができ、クロマ信号の変化の方向性をクラスコードに反映させることができ、クロマ信号の値のみならず、その変化の方向を考慮したクラス分類処理が可能となる。
【０１１７】
また、以上の例においては、コンポジット映像信号からコンポーネント映像信号を生成するようにしたが、例えば、図２１で示すように、図３のクラス分類適応処理回路２２とマトリクス回路６を合成して、クラス分類適応処理回路１２１を構成し、VIF回路３の出力するコンポジット映像信号から、原色ＲＧＢ信号を直接生成するようにすることもできる。この場合においても、クラス分類適応処理回路１２１の構成は、クラス分類適応処理回路２２と同様であるが、係数メモリ５４に記憶される予測係数を決定する学習処理における生徒画像は、コンポジット信号であるが、教師画像は、原色ＲＧＢ信号となり、決定される予測係数も原色ＲＧＢ信号に対応したものとなる。
【０１１８】
さらに、以上においては、NTSC方式のコンポジット信号を用いた例について説明してきたが、例えば、PAL（Phase Alternating Line）方式のコンポジット信号からコンポーネント信号に変換する場合においても、同様の構成で実現させることができる。
【０１１９】
以上によれば、Ｙ／Ｃ分離、色復調、マトリクス処理、画像情報変換処理をクラス分類適応処理により一括して処理することができるようになるため装置規模を小さくし、コストの低減を図ることが可能になると共に、コンポジット信号からコンポーネント信号への変換における画質劣化、特に、画質劣化が生じやすいエッジの方向を正しく区別したクラスコードの分類ができないことによる画質劣化を低減させることが可能となる。
【０１２０】
上述した一連の処理は、ハードウェアにより実行させることもできるが、ソフトウェアにより実行させることもできる。一連の処理をソフトウェアにより実行させる場合には、そのソフトウェアを構成するプログラムが、専用のハードウェアに組み込まれているコンピュータ、または、各種のプログラムをインストールすることで、各種の機能を実行させることが可能な、例えば汎用のパーソナルコンピュータなどに記録媒体からインストールされる。
【０１２１】
図２２は、テレビジョン受像機、または、学習装置をそれぞれソフトウェアにより実現する場合のパーソナルコンピュータの実施の形態の構成を示している。パーソナルコンピュータのCPU２０１は、パーソナルコンピュータの動作の全体を制御する。また、CPU２０１は、バス２０４および入出力インタフェース２０５を介してユーザからキーボードやマウスなどからなる入力部２０６から指令が入力されると、それに対応してROM(Read Only Memory)２０２に格納されているプログラムを実行する。あるいはまた、CPU２０１は、ドライブ２１０に接続された磁気ディスク２１１、光ディスク２１２、光磁気ディスク２１３、または半導体メモリ２１４から読み出され、記憶部２０８にインストールされたプログラムを、RAM(Random Access Memory)２０３にロードして実行する。これにより、上述した画像処理装置の機能が、ソフトウェアにより実現されている。さらに、CPU２０１は、通信部２０９を制御して、外部と通信し、データの授受を実行する。
【０１２２】
プログラムが記録されている記録媒体は、図２２に示すように、コンピュータとは別に、ユーザにプログラムを提供するために配布される、プログラムが記録されている磁気ディスク２１１（フレキシブルディスクを含む）、光ディスク２１２（CD-ROM(Compact Disc-Read Only Memory)，DVD（Digital Versatile Disc）を含む）、光磁気ディスク２１３（MD（Mini-Disc）を含む）、もしくは半導体メモリ２１４などよりなるパッケージメディアにより構成されるだけでなく、コンピュータに予め組み込まれた状態でユーザに提供される、プログラムが記録されているROM２０２や、記憶部２０８に含まれるハードディスクなどで構成される。
【０１２３】
尚、本明細書において、記録媒体に記録されるプログラムを記述するステップは、記載された順序に沿って時系列的に行われる処理は、もちろん、必ずしも時系列的に処理されなくとも、並列的あるいは個別に実行される処理を含むものである。
【０１２４】
【発明の効果】
本発明の画像情報変換装置および方法、並びにプログラムによれば、着目画素の画素信号、および、着目画素の周辺の画素信号を入力コンポジット信号から所定の時空間方向のそれぞれについて輝度信号とクロマ信号とに分離し、着目画素の画素信号、および、着目画素の周辺の画素信号毎に、異なる時空間方向で分離された輝度信号間、または、異なる時空間方向で分離されたクロマ信号間の差分を演算し、演算結果をそれぞれ所定の閾値と比較し、比較結果からなるパターンを生成し、パターン毎に、着目画素を複数のクラスのうちの１つに分類し、入力コンポジット信号から着目画素を変換するための複数の画素の画素信号を抽出し、複数の画素の画素信号と、分類されたクラスに設定されている係数との積和演算により、着目画素の画素信号を変換するようにした。
【０１２６】
本発明の係数算出装置および方法、並びにプログラムによれば、入力された画像信号から入力コンポジット信号を生成し、着目画素の画素信号、および、着目画素の周辺の画素信号を入力コンポジット信号から所定の時空間方向のそれぞれについて輝度信号とクロマ信号とに分離し、着目画素の画素信号、および、着目画素の周辺の画素信号毎に、異なる時空間方向で分離された輝度信号間、または、異なる時空間方向で分離されたクロマ信号間の差分を演算し、演算結果をそれぞれ所定の閾値と比較し、比較結果からなるパターンを生成し、パターン毎に、着目画素を複数のクラスのうちの１つに分類し、入力コンポジット信号から着目画素を変換するための複数の画素の画素信号を抽出し、複数の画素の画素信号と入力された画像信号とを用いた正規方程式より、クラス毎に係数を算出するようにした。
【０１２７】
いずれにおいても、結果として、装置規模を小さくすることができ、コストの低減を図ることが可能になると共に、コンポジット信号からコンポーネント信号への変換における画質劣化、特に、クラスコードの分類処理の誤りが原因とされる画質劣化を低減させることが可能となる。
【図面の簡単な説明】
【図１】従来のテレビジョン受像機の構成例を示すブロック図である。
【図２】従来のテレビジョン受像機の構成例を示すブロック図である。
【図３】本発明を適用したテレビジョン受像機の一実施の形態の構成を示すブロック図である。
【図４】図１のクラス分類適応処理回路の構成例を示すブロック図である。
【図５】図４のクラスコード決定部の構成例を示すブロック図である。
【図６】図３のテレビジョン受像機の表示処理を説明するフローチャートである。
【図７】クラスタップの例を示す図である。
【図８】パターン検出処理を説明するフローチャートである。
【図９】簡易Ｙ／Ｃ分離処理を説明する図である。
【図１０】簡易Ｙ／Ｃ分離処理を説明する図である。
【図１１】簡易Ｙ／Ｃ分離処理を説明する図である。
【図１２】クラスコード決定処理を説明するフローチャートである。
【図１３】予測タップの例を示す図である。
【図１４】着目画素上に水平方向のエッジがある場合を説明する図である。
【図１５】着目画素上に垂直方向のエッジがある場合を説明する図である。
【図１６】着目画素に動きがある場合を説明する図である。
【図１７】着目画素上に垂直方向のエッジがある場合を説明する図である。
【図１８】学習装置を説明するブロック図である。
【図１９】学習装置の学習処理を説明するフローチャートである。
【図２０】クラスコード決定部のその他の構成例を示すブロック図である。
【図２１】テレビジョン受像機のその他の構成例を示すブロック図である。
【図２２】媒体を説明する図である。
【符号の説明】
２２ クラス分類適応処理回路， ５１ 領域抽出部， ５２ パターン検出部， ５３ クラスコード決定部， ５４ 係数メモリ， ５５ 領域抽出部，５６ 予測演算部， ７１ 減算器， ７２ ABS回路， ７３ 比較器， ９１ NTSCエンコーダ， ９２ 領域抽出部， ９３ パターン検出部， ９４クラスコード決定部， ９５ 領域抽出部， ９６ 正規方程式生成部， ９７ 係数決定部， ９８ メモリ， １１１ 減算器， １１２ 比較器， １２１ クラス分類適応処理回路[0001]
BACKGROUND OF THE INVENTION
  The present inventionImage information conversion device, coefficient calculation deviceAnd a method, a storage device, a recording medium, and a program, in particular, it is possible to suppress deterioration in image quality that occurs when converting a composite signal into a component signal.Image information conversion device, coefficient calculation deviceAnd a method, a storage device, a recording medium, and a program.
[0002]
[Prior art]
Generally, in a television receiver, a composite signal of a video signal is converted into a component signal and displayed as a video. As video signals, NTSC (National Television Standards Committee) type composite video signals are widely used.
[0003]
FIG. 1 shows a configuration example of a conventional television receiver. The tuner 2 demodulates the signal received by the antenna 1 and outputs it to the video intermediate frequency signal processing circuit (VIF circuit) 3. The composite video signal processed and output by the VIF circuit 3 is input to a Y / C (Y: luminance signal, C: chroma signal) separation circuit 4. The Y / C separation circuit 4 separates the luminance signal Y and the chroma signal C from the input composite video signal, and outputs them to the matrix circuit 6 and the chroma demodulation circuit 5, respectively. The chroma demodulation circuit 5 demodulates the input chroma signal C, generates a color difference signal RY and a color difference signal BY, and supplies them to the matrix circuit 6. The matrix circuit 6 generates primary color RGB signals from the input luminance signal Y and the color difference signals RY and BY, and outputs them to the display device 7.
[0004]
Next, the operation will be described. The tuner 2 receives the radio wave of the broadcasting station of the channel designated by the user via the antenna 1 and outputs the demodulated signal to the VIF circuit 3. The VIF circuit 3 processes the signal output from the tuner 2 and outputs an NTSC composite video signal to the Y / C separation circuit 4. The Y / C separation circuit 4 separates the luminance signal Y and the chroma signal C from the composite video signal, and outputs the luminance signal Y to the matrix circuit 6 and the chroma signal C to the chroma demodulation circuit 5.
[0005]
The chroma demodulation circuit 5 demodulates the chroma signal C input from the Y / C separation circuit 4 to generate a color difference signal RY and a color difference signal BY. The matrix circuit 6 synthesizes the luminance signal Y supplied from the Y / C separation circuit 4 and the color difference signals RY and BY input from the chroma demodulation circuit 5 to generate primary color RGB signals for display. Output to device 7.
[0006]
FIG. 2 shows another configuration example of a conventional television receiver. The basic configuration is the same as that of the television receiver of FIG. 1, but in FIG. 2, a pixel number conversion circuit 11 is provided. The pixel number conversion circuit 11 changes the pixel density based on the luminance signal Y supplied from the Y / C separation circuit 4 and the color difference signals RY and BY input from the chroma demodulation circuit 5, so-called IP conversion (Interlace / Progressive conversion) or the like is executed and output to the matrix circuit 6. 2, pixel density conversion processing and IP conversion processing are added to the operation of the television receiver in FIG. 1 by the pixel number conversion circuit 11. .
[0007]
[Problems to be solved by the invention]
As described above, in the conventional television receiver, the composite video signal is first separated into the luminance signal Y and the chroma signal C by the Y / C separation circuit 4, and then the chroma signal C is demodulated to obtain the base signal. The signal is converted into a component signal composed of a band luminance signal Y and color difference signals RY and BY. The matrix circuit 6 further generates primary color RGB signals from the component signals. For this reason, there is a problem that not only the circuit configuration becomes complicated, but also the scale becomes large and the cost increases.
[0008]
Further, as a conventional Y / C separation method, a filter composed of a two-dimensional Y / C separation circuit, a three-dimensional Y / C separation circuit, etc. has been proposed. There has been a problem that image quality deterioration easily occurs due to an error of / C separation.
[0009]
In order to solve these problems, a method using a classification adaptation process has been proposed. In this method using the class classification adaptive processing, in order to obtain component signals such as the luminance signal Y and the color difference signals RY and BY at the target pixel, first, they are arranged in the vicinity including the target pixel. A feature quantity of a pixel (this pixel is called a class tap) is obtained from the composite signal, and a class classified in advance for each feature quantity is specified.
[0010]
Then, for each class, a pixel composed of a composite signal of neighboring pixels including the pixel of interest using a preset coefficient (this pixel is called a prediction tap and may be the same as the class tap) ) To obtain the component signal of the pixel of interest directly from the composite signal.
[0011]
It has been confirmed that image quality degradation due to Y / C separation errors such as dot interference and cross color occurs in the edge portion of the image (where the pixel value between adjacent pixels changes rapidly). The amount is preferably such that an edge portion can be detected.
[0012]
By the way, since the NTSC composite signal is a signal in which the chroma signal C is balanced-modulated with the luminance signal Y and multiplexed, the change point cannot be correctly captured with the signal as it is. Therefore, the luminance signal Yh by the simple Y / C separation process in the horizontal direction obtained from the pixel of interest and the antiphase pixel existing in the left and right direction, the vertical signal obtained from the pixel of interest and the antiphase pixel existing in the vertical direction thereof. Luminance signal Yv by simple Y / C separation processing and luminance signal by simple Y / C separation processing in the time direction obtained from pixels existing at the same position two fields before or after the field where the pixel of interest exists. It is considered that the feature value of the pixel of interest is obtained by obtaining Yt, obtaining the absolute value of the difference between each other, comparing each with a predetermined threshold value, and converting it into binarized information from the magnitude relationship with the threshold value. Yes. That is, for example, | brightness signal Yh−brightness signal Yv | is obtained, and converted to 1 when it is larger than the threshold and 0 when it is equal to or smaller than the threshold from comparison with a predetermined threshold. In this way, when it is larger than the threshold value, it can be confirmed that the horizontal direction or the vertical direction is greatly changed as viewed from the target pixel.
[0013]
However, with the method as described above, for example, there is no difference between when the edge exists in the horizontal direction on the target pixel and when the edge exists in the vertical direction. Then, as a matter of fact, although the direction in which the edge is generated is different, the target pixel in which the edge exists in the horizontal direction on the target pixel and the edge in the vertical direction on the target pixel exist. As a result, the pixel of interest cannot be classified correctly, and as a result, when the composite signal is converted into the component signal, image quality degradation such as dot interference and cross color as described above occurs. was there.
[0014]
The present invention has been made in view of such circumstances, and has a simple and small circuit configuration, and further, degradation of image quality that occurs during conversion from a composite signal to a component signal. It is intended to suppress image quality degradation that occurs when the pixel classification method is not appropriate.
[0015]
[Means for Solving the Problems]
  The image information conversion apparatus of the present invention is a pixel signal of a pixel of interest., And pixel signals around the pixel of interestFrom the input composite signalSeparate into luminance signal and chroma signal for each given spatio-temporal directionSeparating means;The difference between luminance signals separated in different spatiotemporal directions or chroma signals separated in different spatiotemporal directions by the separating means for each pixel signal of the target pixel and pixel signals around the pixel of interest. Difference calculation means for calculation and calculation results of difference calculation meansComparing means for comparing each with a predetermined threshold, and the comparison result by the comparing meansConsist ofPattern generation means for generating a pattern and patternEveryClass classification means for classifying the pixel of interest into one of a plurality of classes and an input composite signalFor converting the pixel of interestExtraction means for extracting pixel signals of a plurality of pixels, pixel signals of a plurality of pixels, and classes classified by the class classification meansIs set toWith coefficientSum of productsAnd pixel-of-interest signal conversion means for converting a pixel signal of the pixel of interest by calculation.
[0016]
  The separating means is, Out of phase with the pixel of interest, andOut of space-time direction3 pixels including the pixel of interest added to the 2 pixels closest in the left-right direction, in phase with the pixel of interest, andOut of space-time direction3 pixels obtained by adding the pixel of interest to the 2 pixels closest in the vertical direction, and in phase opposite to the pixel of interest, andOut of space-time directionEach pixel signal of two pixels, which is the pixel closest to the time direction plus the pixel of interestUsing, Each pixel signal of the pixel of interestIn addition to separating the luminance signal and chroma signal in the horizontal direction, vertical direction, and time direction,In anti-phase with pixels around the pixel of interest, andOut of space-time direction3 pixels obtained by adding the peripheral pixels of the target pixel to the two closest pixels in the left-right direction, in reverse phase with the peripheral pixels of the target pixel, andOut of space-time direction3 pixels obtained by adding the peripheral pixels of the pixel of interest to the two pixels closest to the vertical direction, and an opposite phase to the pixels around the pixel of interest, andOut of space-time directionEach pixel signal of two pixels obtained by adding pixels around the pixel of interest to the one pixel closest in the time directionUsing, The pixel signals of the pixels around the pixel of interestLuminance and chroma signals in the left / right, up / down, and time directionsIt can be made to separate.
[0017]
  Pixels around the pixel of interest are in opposite phase to the pixel of interest and in the left-right directionPixels closest to the vertical direction or time directionAnd so on.
[0018]
  SaidThe difference calculation means includes a difference between the luminance signal separated in the left-right direction and the luminance signal separated in the up-down direction and the luminance signal separated in the up-down direction for each pixel signal of the pixel of interest and the pixel signals around the pixel of interest. And the difference between luminance signals separated in the time direction, and the difference between the luminance signal separated in the time direction and the luminance signal separated in the left-right direction, or between the chroma signal separated in the left-right direction and the chroma signal separated in the vertical direction Difference between the chroma signal separated in the vertical direction and the chroma signal separated in the time direction, and the difference between the chroma signal separated in the time direction and the luminance chroma separated in the left-right direction can be calculated. .
[0019]
  Difference between luminance signals separated in different spatiotemporal directions or chroma signals separated in different spatiotemporal directions for each pixel signal of the pixel of interest and pixel signals around the pixel of interest by the difference calculation meansThe absolute value ofAbsolute value calculation meansThe comparison means includes:As the calculation result of the difference calculation means, the pixel signal of the target pixel and the luminance signal separated in different spatiotemporal directions or the chroma signal separated in different spatiotemporal directions for each pixel signal around the target pixel Difference betweenThe absolute value of each can be compared with a predetermined threshold value.
[0020]
  The pixel-of-interest signal conversion means includes pixel signals of a plurality of pixels and a class classified by the class classification means.Set for eachWith coefficientSum of productsBy calculation, the pixel of interestPixel signal is luminance signal or chroma signalCan be converted to
[0021]
  The pixel-of-interest signal conversion means includes pixel signals of a plurality of pixels and a class classified by the class classification means.Set for eachWith coefficientSum of productsBy calculation, the pixel signal of the pixel of interest can be converted into a component signal.
Display means for displaying the pixel signal of the pixel of interest converted into the component signal can be further provided.
[0022]
  The image information conversion method of the present invention is a pixel signal of a pixel of interest., And pixel signals around the pixel of interestFrom the input composite signalSeparate into luminance signal and chroma signal for each given spatio-temporal directionA separation step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the processing of the separation step Difference calculation step for calculating the difference, and calculation result in the difference calculation step processingEach of the comparison step and the comparison result of the comparison step processingConsist ofA pattern generation step for generating a pattern and a patternEveryA class classification step for classifying the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestExtraction step for extracting pixel signals of a plurality of pixels, pixel signals of a plurality of pixels, and classes classified by the processing of the class classification stepIs set toWith coefficientSum of productsA pixel-of-interest signal conversion step of converting a pixel signal of the pixel of interest by calculation.
[0023]
  The program of the first recording medium of the present invention is the pixel signal of the pixel of interest, And pixel signals around the pixel of interestFrom the input composite signalControls separation into luminance and chroma signals for each given spatiotemporal directionA separation control step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the separation control step processing Difference calculation control step for controlling the calculation of the difference between the two, and the calculation result in the processing of the difference calculation control stepA comparison control step for controlling comparison with each of the predetermined threshold values, and a comparison result in the processing of the comparison control stepConsist ofPattern generation control step for controlling pattern generation, and patternEveryA class classification control step for controlling classification of the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestAn extraction control step for controlling extraction of pixel signals of a plurality of pixels, a pixel signal of a plurality of pixels, and a class whose classification is controlled by the processing of the class classification control stepIs set toWith coefficientSum of productsAnd a pixel-of-interest signal conversion control step for controlling the conversion of the pixel signal of the pixel of interest by calculation.
[0024]
  The first program of the present invention is a pixel signal of a pixel of interest., And pixel signals around the pixel of interestFrom the input composite signalControls separation into luminance and chroma signals for each given spatiotemporal directionA separation control step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the separation control step processing Difference calculation control step for controlling the calculation of the difference between the two, and the calculation result in the processing of the difference calculation control stepA comparison control step for controlling comparison with each of the predetermined threshold values, and a comparison result in the processing of the comparison control stepConsist ofPattern generation control step for controlling pattern generation, and patternEveryA class classification control step for controlling classification of the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestAn extraction control step for controlling extraction of pixel signals of a plurality of pixels, a pixel signal of a plurality of pixels, and a class whose classification is controlled by the processing of the class classification control stepIs set toWith coefficientSum of productsThe calculation causes the computer to execute a target pixel signal conversion control step for controlling the conversion of the pixel signal of the target pixel.
[0029]
  The coefficient calculation apparatus according to the present invention includes a composite signal generation unit that generates an input composite signal from an input image signal, and a pixel signal of a pixel of interest., And pixel signals around the pixel of interestFrom the input composite signalSeparate into luminance signal and chroma signal for each given spatio-temporal directionSeparating means;The difference between luminance signals separated in different spatiotemporal directions or chroma signals separated in different spatiotemporal directions by the separating means for each pixel signal of the target pixel and pixel signals around the pixel of interest. Difference calculation means for calculation and calculation results of difference calculation meansComparing means for comparing each with a predetermined threshold, and the comparison result by the comparing meansConsist ofPattern generation means for generating a pattern and patternEveryClass classification means for classifying the pixel of interest into one of a plurality of classes and an input composite signalFor converting the pixel of interestExtraction means for extracting pixel signals of a plurality of pixels, pixel signals of a plurality of pixels, and input image signalsFrom the normal equation using and,classEveryAnd a calculating means for calculating a coefficient.
[0030]
  The storage device of the present invention is as claimed.12The coefficient calculated by the coefficient calculation apparatus described in the above is stored.
[0031]
  The coefficient calculation method of the present invention includes a composite signal generation step for generating an input composite signal from an input image signal, and a pixel signal of a pixel of interest., And pixel signals around the pixel of interestFrom the input composite signalSeparate into luminance signal and chroma signal for each given spatio-temporal directionA separation step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the processing of the separation step Difference calculation step for calculating the difference, and calculation result in the difference calculation step processingEach of the comparison step and the comparison result of the comparison step processingConsist ofA pattern generation step for generating a pattern and a patternEveryA class classification step for classifying the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestAn extraction step for extracting pixel signals of a plurality of pixels, a pixel signal of the plurality of pixels, and an input image signalFrom the normal equation using and,classEveryAnd a calculating step for calculating a coefficient.
[0032]
  First of the present invention2The recording medium program includes a composite signal generation control step for controlling generation of an input composite signal from an input image signal, and a pixel signal of a target pixel., And pixel signals around the pixel of interestInput composite signal fromControls separation into luminance and chroma signals for each given spatiotemporal directionA separation control step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the separation control step processing Difference calculation control step for controlling the calculation of the difference between the two, and the calculation result in the processing of the difference calculation control stepA comparison control step for controlling comparison with each of the predetermined threshold values, and a comparison result in the processing of the comparison control stepConsist ofPattern generation control step for controlling pattern generation, and patternEveryA class classification control step for controlling classification of the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestExtraction control step for controlling extraction of pixel signals of a plurality of pixels, pixel signals of a plurality of pixels, and input image signalsFrom the normal equation using and,classEveryAnd a calculation control step for controlling calculation of the coefficient.
[0033]
  First of the present invention2The program includes a composite signal generation control step for controlling generation of an input composite signal from an input image signal, and a pixel signal of a target pixel., And pixel signals around the pixel of interestInput composite signal fromControls separation into luminance and chroma signals for each given spatiotemporal directionA separation control step;For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, between the luminance signals separated in different spatiotemporal directions or between chroma signals separated in different spatiotemporal directions by the separation control step processing Difference calculation control step for controlling the calculation of the difference between the two, and the calculation result in the processing of the difference calculation control stepA comparison control step for controlling comparison with each of the predetermined threshold values, and a comparison result in the processing of the comparison control stepConsist ofPattern generation control step for controlling pattern generation, and patternEveryA class classification control step for controlling classification of the pixel of interest into one of a plurality of classes, and an input composite signalFor converting the pixel of interestExtraction control step for controlling extraction of pixel signals of a plurality of pixels, pixel signals of a plurality of pixels, and input image signalsFrom the normal equation using and,classEveryA calculation control step for controlling calculation of the coefficient is executed by a computer.
[0034]
  In the image information conversion apparatus and method and program of the present invention, the pixel signal of the pixel of interest, And pixel signals around the pixel of interestFrom the input composite signalSeparated into luminance signal and chroma signal for each given spatio-temporal direction,The difference between the luminance signals separated in different spatiotemporal directions or the chroma signals separated in different spatiotemporal directions is calculated and calculated for each pixel signal of the pixel of interest and the pixel signals around the pixel of interest. resultAre compared with a predetermined threshold value, and the comparison resultConsist ofPattern is generated, patternEveryThe target pixel is classified into one of a plurality of classes, and from the input composite signalFor converting the pixel of interestPixel signals of multiple pixels are extracted, pixel signals of multiple pixels, and classified classesIs set toWith coefficientSum of productsThe pixel signal of the pixel of interest is converted by the calculation.
[0036]
  In the coefficient calculation apparatus and method, and the program of the present invention, an input composite signal is generated from an input image signal, and a pixel signal of a pixel of interest, And pixel signals around the pixel of interestFrom the input composite signalSeparated into luminance signal and chroma signal for each given spatio-temporal direction,The difference between the luminance signals separated in different spatiotemporal directions or the chroma signals separated in different spatiotemporal directions is calculated and calculated for each pixel signal of the pixel of interest and the pixel signals around the pixel of interest. resultAre compared with a predetermined threshold value, and the comparison resultConsist ofPattern is generated, patternEveryThe target pixel is classified into one of a plurality of classes, and from the input composite signalFor converting the pixel of interestImage signals that are extracted from pixel signals of multiple pixels and input as pixel signals of multiple pixelsFrom the normal equation with and,classEveryA coefficient is calculated.
[0037]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 3 is a block diagram showing the configuration of an embodiment of a television receiver to which the present invention is applied, in which parts corresponding to those in FIGS. 1 and 2 are given the same reference numerals, Description is omitted as appropriate.
[0038]
In this configuration, an A / D (Analog / Digital) converter 21 and a class classification adaptive processing circuit 22 are arranged between the VIF circuit 3 and the matrix circuit 6. The A / D converter 21 converts the NTSC composite video signal composed of the analog signal input from the VIF circuit 3 into a digital signal and outputs the digital signal to the class classification adaptive processing circuit 22. The class classification adaptive processing circuit 22 is a component composed of a luminance signal Y and color difference signals RY and BY signals from an NTSC composite video signal converted into a digital signal input from the A / D converter 21. The signal is generated directly by the classification adaptation process. A prediction coefficient for predicting and generating a component signal from a composite video signal is generated by learning using a component signal as a teacher image and an NTSC composite video signal created by NTSC modulation of the component signal as a student image. Then, a component signal is generated by performing mapping (prediction calculation processing) using the prediction coefficient.
[0039]
FIG. 4 shows a configuration example of the class classification adaptive processing circuit 22. The NTSC composite video signal input from the VIF circuit 3 is supplied to the region extraction units 51 and 55. The area extraction unit 51 extracts pixels (class taps) necessary for class classification from the composite video signal converted into the input digital signal, and outputs the extracted pixels to the pattern detection unit 52. The pattern detection unit 52 detects the pattern of the composite video signal based on the input class tap. This pattern is obtained by Y / C separation processing in the horizontal direction, the vertical direction, and the time direction, respectively, for the pixel of interest and the three pixels that are closest to the left and right of the pixel of interest and in opposite phase to the pixel of interest. Value. The pattern detection process will be described later.
[0040]
The class code determination unit 53 determines a class code based on the pattern detected by the pattern detection unit 52 and outputs the class code to the coefficient memory 54. That is, the class code determination unit 53 performs Y / C separation in the vertical direction and the chroma signal Ch obtained by performing Y / C separation in the horizontal direction with respect to the signal of interest obtained by the pattern detection unit 52 and the pixels having opposite left and right phases. Chroma signal Cv, Y / C separated chroma signal Cv in the vertical direction and chroma signal Ct separated in the Y / C direction in the time direction, and chroma signal Ct separated in the Y / C direction in the time direction in the Y / C separation in the horizontal direction The absolute value of the mutual difference is obtained from the combination of the chroma signals Ch, compared with a predetermined threshold value, a class code is determined based on the comparison result, and is output to the coefficient memory 54. The detailed configuration of the class code determination unit 53 will be described later with reference to FIG.
[0041]
The coefficient memory 54 stores the prediction coefficient for each class obtained by learning in advance, reads the prediction coefficient corresponding to the class code input from the class code determination unit 53, and outputs the prediction coefficient to the prediction calculation unit 56. Note that the prediction coefficient learning process of the coefficient memory 54 will be described later with reference to the block diagram of the learning apparatus in FIG.
[0042]
The region extraction unit 55 extracts pixels (prediction taps) necessary for predictive generation of component signals from the composite video signal input from the VIF circuit 3, and outputs them to the prediction calculation unit 56. The prediction calculation unit 56 multiplies the prediction tap input from the region extraction unit 55 by the prediction coefficient input from the coefficient memory 54 to generate a luminance signal Y that is one of the component signals.
[0043]
More specifically, the coefficient memory 54 outputs a prediction coefficient corresponding to the class code supplied from the class code determination unit 53 to the prediction calculation unit 56. The prediction calculation unit 56 uses the prediction tap extracted from the pixel value at a predetermined pixel position of the NTSC composite video signal supplied from the region extraction unit 55 and the prediction coefficient supplied from the coefficient memory 54, A component signal is obtained (predicted and estimated) by executing a product-sum operation shown in the following equation (1).
[0044]
y = w1 * x1 + w2 * x2 ++ ... + wn * xn (1)
Here, y is a pixel of interest. In addition, x1,..., Xn are prediction taps, and w1,.
[0045]
In addition, although illustration is abbreviate | omitted, the circuit for producing | generating other color difference signal RY and color difference signal BY among component signals is also comprised similarly. The configuration is the same as in the case shown in FIG. 4, and the coefficients stored in the coefficient memory 54 are prediction coefficients for generating the luminance signal Y in the case shown in FIG. When the signal RY or BY is generated, a prediction coefficient for generating the color difference signal RY or BY is stored.
[0046]
Next, a detailed configuration of the class code determination unit 53 will be described with reference to FIG. The subtracting unit 71 of the class code determining unit 53 performs the Y / C separation processing by the pattern detecting unit 52 and the horizontal and vertical directions and the vertical direction of the chroma signal C of pixels of opposite phase on the left and right sides thereof. Time direction and difference between time direction and horizontal direction (horizontal direction and vertical direction (Ch-Cv), vertical direction and time direction (Cv-Ct), and time direction and horizontal direction (Ct-Ch)) Is output to an ABS (Absolute) circuit 72.
[0047]
The ABS circuit 72 receives the signal of interest input from the subtracting unit 71, the chroma signal Ch of Y / C separation in the horizontal direction and the chroma signal Cv of Y / C separation in the vertical direction of the left and right anti-phase pixels, vertical A chroma signal Cv separated by Y / C in the direction and a chroma signal Ct separated by Y / C in the time direction, and a chroma signal Ct separated by Y / C in the time direction and a chroma signal Ch separated by Y / C in the horizontal direction, respectively. Absolute values (horizontal and vertical | Ch-Cv |, vertical and temporal | Cv-Ct |, and temporal and horizontal | Ct-Ch |) are obtained and output to the comparator 73. The comparator 73 compares the absolute value of each difference with a predetermined threshold value. When each absolute value is larger than the predetermined threshold value, 1 is set. When each absolute value is less than the predetermined threshold value, 0 is set. By doing so, binarized information is generated and output to the coefficient memory 54 as a class code.
[0048]
Next, a process in which the television receiver shown in FIG. 3 converts the composite video signal into a component signal and displays it on the display device 7 will be described with reference to the flowchart of FIG.
[0049]
In step S 1, the tuner 2 receives a signal of a predetermined frequency via the antenna 1, demodulates it, and outputs it to the VIF circuit 3. In step S <b> 2, the VIF circuit 3 generates an NTSC composite signal from the input signal and outputs the composite signal to the A / D converter 21. In step S <b> 3, the A / D converter 21 converts the input composite signal from an analog signal to a digital signal, and outputs it to the class classification adaptive processing circuit 22.
[0050]
In step S4, the region extraction unit 51 extracts class taps. Here, the class tap will be described with reference to FIG. When the NTSC composite video signal input to the region extraction unit 51 is sampled at four times the color subcarrier frequency fsc (frequency for sub-carrier), each pixel is detected as a four-phase signal as shown in FIG. . Here, a black circle indicates a phase angle of 0 °, a white square indicates a phase angle of 90 °, a white circle indicates a phase angle of 180 °, and a black square indicates a phase angle of 270 °. As shown in FIG. 7, the phase of each pixel changes by 90 degrees in the horizontal direction, and the phase changes by 180 degrees in the vertical direction. This occurs because the composite signal is a signal in which the luminance signal Y and the I signal or Q signal are superimposed. That is, there are four signals of Y + I, Y + Q, Y-I, and Y-Q, and these signals are generated when they are arranged in the horizontal direction in the same arrangement order. In the vertical direction, a column is formed for each signal superimposed on the luminance signal Y (for each I signal and Q signal).
[0051]
Further, each phase changes by 180 degrees in the time direction. In the figure, for example, the pixel P1 on the field # 0 has a phase angle of 0 degree, whereas the phase P is 1 in the field # 0. The pixel P13 existing at the same position on the field # -2 before the frame (one frame is composed of two fields and two fields before) has a phase angle of 180 degrees.
[0052]
The field has a top field and a bottom field, which are alternately arranged, and one frame is formed from the top field and the bottom field. In both the top field and the bottom field, pixels are arranged for each line, but in the top field and the bottom field, the line in which the pixels are arranged is shifted by one line. For this reason, the field including the pixel of interest and the field in which the pixel is arranged at the same position are fields that are separated by an even number field from the field including the pixel of interest. For example, as shown in FIG. 7, when the target pixel exists on the field # 0, the pixel is arranged at the same position on the field # -2 two fields before.
[0053]
Here, for example, when the target pixel is the pixel P1 on the field # 0 in FIG. 7, the region extracting unit 51 performs the pixels P1, P2, P3, P11, P12, P21, P22, P23, P31, P32, and P33. 14 pixels of the pixels P13, P24, and P34 on the field # -2 are extracted as class taps and output to the pattern detection unit 52. These class taps exist at the closest position in the right direction of the pixel of interest P1 on the field # 0, the pixels P2, P3, P11, and P12 having the opposite phases closest to the top, bottom, left, and right, and the pixel of interest P1. An anti-phase pixel P2, an anti-phase pixel P1, P21, P22, P23 closest to the up / down / left / right direction, an anti-phase pixel P3 present at the closest position in the left direction of the pixel of interest P1; This is a total of 14 taps including the pixels P1, P31, P32, and P33 that are closest in the vertical and horizontal directions, the pixels P1 to P3, and the pixels P13, P24, and P34 on the previous field # -2 in the time direction.
[0054]
In step S5, the pattern detection unit 52 executes pattern detection processing based on the input class tap. Here, the pattern detection processing of the pattern detection unit 52 will be described with reference to the flowchart of FIG.
[0055]
In step S <b> 41, the pattern detection unit 52 performs simple Y / C separation in the vertical direction between the pixel of interest and the pixel of the opposite phase existing at the closest position on the left and right. That is, for example, with respect to the target pixel, as shown in FIG. 9, the target pixel P1 is set in the vertical direction by the following formulas (2) and (3) using the pixel values of the pixels P11 and P12 above and below the target pixel P1. To obtain the chroma signal Cv1 and the luminance signal Yv1.
[0056]
Cv1 = 0.5 * P1-0.25 * P11-0.25 * P12 (2)
Yv1 = 0.5 × P1 + 0.25 × P11 + 0.25 × P12 (3)
[0057]
Here, Cv1 indicates a chroma signal subjected to simple Y / C separation in the vertical direction of the target pixel P1, and Yv1 indicates a luminance signal subjected to simple Y / C separation in the vertical direction of the target pixel P1.
[0058]
Similarly, the anti-phase pixels P2 and P3 present at the closest positions on the left and right of the pixel of interest are simply Y / C separated in the vertical direction from the pixels P2 and P3 by the following equations (4) to (7). Then, chroma signals Cv2, Cv3 and luminance signals Yv2, Yv3 are obtained.
[0059]
Cv2 = 0.5 × P2−0.25 × P21−0.25 × P22 (4)
Yv2 = 0.5 × P2 + 0.25 × P21 + 0.25 × P22 (5)
Cv3 = 0.5 * P3-0.25 * P31-0.25 * P32 (6)
Yv3 = 0.5 × P3 + 0.25 × P31 + 0.25 × P32 (7)
[0060]
In step S <b> 42, the pattern detection unit 52 performs simple Y / C separation in the horizontal direction between the pixel of interest and the pixel of the opposite phase existing at the closest position on the left and right. That is, for example, with respect to the target pixel, as shown in FIG. 10, the target pixel P1 is moved in the horizontal direction by the following formulas (8) and (9) using the pixel values of the pixels P2 and P3 above and below the target pixel P1. To obtain a chroma signal Ch1 and a luminance signal Yh1.
[0061]
Ch1 = 0.5 × P1-0.25 × P2-0.25 × P3 (8)
Yh1 = 0.5 × P1 + 0.25 × P2 + 0.25 × P3 (9)
[0062]
Here, Ch1 indicates a chroma signal obtained by simple Y / C separation in the horizontal direction of the target pixel P1, and Yh1 indicates a luminance signal obtained by simple Y / C separation in the horizontal direction of the target pixel P1.
[0063]
Similarly, the anti-phase pixels P2 and P3 present at the closest positions on the left and right of the pixel of interest are simply Y / C separated in the horizontal direction from the pixels P2 and P3 by the following equations (10) to (13). Then, chroma signals Ch2 and Ch3 and luminance signals Yh2 and Yh3 are obtained.
[0064]
Ch2 = 0.5 × P2−0.25 × P23−0.25 × P1 (10)
Yh2 = 0.5 × P2 + 0.25 × P23 + 0.25 × P1 (11)
Ch3 = 0.5 × P3-0.25 × P1-0.25 × P33 (12)
Yh3 = 0.5 × P3 + 0.25 × P1 + 0.25 × P33 (13)
[0065]
In step S <b> 43, the pattern detection unit 52 performs simple Y / C separation in the time direction between the pixel of interest and the pixel of the opposite phase existing at the closest position on the left and right. That is, for example, with respect to the target pixel, as shown in FIG. 11, the following equation (14) is used by using the pixel values of the target pixel P1 existing on the field # 0 and the pixel P13 on the field # -2 immediately before the target pixel P1. ) And Expression (15), the target pixel P1 is simply Y / C separated in the time direction, and a chroma signal Ct1 and a luminance signal Yt1 are obtained.
[0066]
Ct1 = 0.5 × P1-0.5 × P13 (14)
Yt1 = 0.5 × P1 + 0.5 × P13 (15)
[0067]
Here, Ct1 indicates a chroma signal subjected to simple Y / C separation in the time direction of the target pixel P1, and Yt1 indicates a luminance signal subjected to simple Y / C separation in the time direction of the target pixel P1.
[0068]
Similarly, the anti-phase pixels P2 and P3 existing at the closest positions on the left and right of the pixel of interest are simply Y / C separated in the time direction by the following equations (16) to (19). Then, chroma signals Ct2 and Ct3 and luminance signals Yt3 and Yt3 are obtained.
[0069]
Ct2 = 0.5 × P2-0.5 × P24 (16)
Yt2 = 0.5 × P2 + 0.5 × P24 (17)
Ct3 = 0.5 × P3-0.5 × P34 (18)
Yt3 = 0.5 × P3 + 0.5 × P34 (19)
[0070]
In step S44, the pattern detection unit 52 outputs the chroma signals Cv1 to Cv3, Ch1 to Ch3, and Ct1 to Ct3 obtained by the above processing to the class code determination unit 53.
[0071]
The nine chroma signals C generated by the simple Y / C separation processing by the above processing are output to the class code determination unit 53.
[0072]
Now, the description returns to the flowchart of FIG.
[0073]
In step S6, the class code determination unit 53 executes class code determination processing. Here, the class code determination processing will be described with reference to the flowchart of FIG. Here, first, a description will be given of processing for determining a chroma code Cv1 obtained by simple Y / C separation in the vertical direction of the pixel of interest and a class code obtained from Ch1 obtained by simple Y / C separation in the horizontal direction. .
[0074]
In step S51, the subtraction unit 71 obtains a difference (Ch1-Cv1) between the chroma signal Ch1 and the chroma signal Cv1, and outputs the difference to the ABS circuit 72. In step S52, the ABS circuit 72 calculates an absolute value | Ch1-Cv1 | of the difference (Ch1-Cv1) and outputs it to the comparator 73. In step S53, the comparator 73 determines whether or not the input absolute value | Ch1-Cv1 | is larger than a predetermined threshold value. If so, in step S54, the class code for the absolute value | Ch1-Cv1 | code | Ch1-Cv1 | is set as 1. If the absolute value | Ch1-Cv1 | is not larger than the predetermined threshold value in step S53, the comparison unit 73 sets 0 as the class code code | Ch1-Cv1 | for the absolute value | Ch1-Cv1 | in step S55. To do.
[0075]
The class code determination unit 53 performs the same processing with the absolute values | Cv1-Ct1 |, | Ct1-Ch1 |, | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 |, | Ch3-Cv3 |, | Cv3-Ct3 |, | Ct3-Ch3 | are obtained for each, and a 9-bit class code (code | Cv1-Ct1 |, code | Ct1-Ch1 |, code | Ch2-Cv2 |, code | Cv2-Ct2 |, code | Ct2-Ch2 |, code | Ch3-Cv3 |, code | Cv3-Ct3 |, code | Ct3-Ch3 |) are determined and output to the coefficient memory 54.
[0076]
The class code determination unit 53 has the absolute value | Ch1-Cv1 |, | Cv1-Ct1 |, | Ct1-Ch1 |, | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 |, | Ch3-Cv3 |, | Cv3-Ct3 |, | Ct3-Ch3 | are obtained by similar processing. In this case, | Ch1-Cv1 |, | Cv1-Ct1 |, | Ct1-Ch1 |, | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 |, | Ch3-Cv3 |, | Cv3-Ct3 | , | Ct3-Ch3 | may be obtained by repeating the same process as the process shown in the flowchart of FIG. 12 nine times, and the configuration shown in FIG. Corresponding to Cv1-Ct1 |, | Ct1-Ch1 |, | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 |, | Ch3-Cv3 |, | Cv3-Ct3 |, | Ct3-Ch3 | Thus, a total of nine may be provided so that parallel processing can be performed.
[0077]
Now, the description returns to the flowchart of FIG.
[0078]
In step S <b> 7, the coefficient memory 54 reads out the prediction coefficient determined in advance by the learning process based on the class code, and outputs it to the prediction calculation unit 56. In step S <b> 8, the region extraction unit 55 extracts a prediction tap from the composite signal and outputs the prediction tap to the prediction calculation unit 56. For example, as shown in FIG. 13, the prediction tap is composed of 5 pixels in the horizontal direction centered on the pixel of interest on the field # 0, 15 pixels composed of the upper and lower 5 pixels, and the field # -2 immediately before it. A total of 21 taps including 5 pixels in the horizontal direction centering on the pixel at the target pixel position and offset taps for adjusting the deviation of the DC component generated during conversion may be used. Further, the prediction tap is not limited to the tap shown in FIG. 13 and may be another tap.
[0079]
In step S <b> 9, the prediction calculation unit 56 calculates the prediction tap input from the region extraction unit 55 using the prediction coefficient, generates a component signal, and outputs the component signal to the matrix circuit 6.
[0080]
In step S <b> 10, the matrix circuit 6 converts the component signal into an RGB signal and outputs it to the display device 7.
[0081]
In step S11, it is determined whether or not signals have been converted for all pixels. If it is determined that signals have not been converted for all pixels, the process returns to step S4. That is, the processes in steps S4 to S11 are repeated until the signals of all the pixels are converted. If it is determined in step S11 that signals have been processed (converted) for all pixels, the processing ends.
[0082]
With the above processing, the classification processing in consideration of the presence / absence of the edge in the horizontal direction and the movement in the time direction can be performed by the processing in steps S5 and S6 (the flowcharts in FIGS. 8 and 12). That is, for example, as shown in FIG. 14, when an edge exists in the horizontal direction on the target pixel P <b> 1 (straight line part in the figure: there is a part where the upper and lower pixel values change extremely with the edge as a boundary). If the class code is set only with the chroma signal of the pixel of interest (with the simple Y / C separated in the horizontal direction, vertical direction, and time direction), for example, in the horizontal direction (simple Y / C in the horizontal direction) When classifying processing is performed in consideration of the absolute value | Ch1-Cv1 | of the difference between the separated chroma signal Ch1 and the chroma signal Cv1 in the vertical direction (simple Y / C separation in the vertical direction), the difference becomes a threshold value. It exceeds th, and the presence of an edge can be recognized. However, even when there is an edge in the vertical direction on the target pixel P1 as shown in FIG. 15, the absolute value | Ch1-Cv1 | exceeds the threshold th, and the pixel P1 is displayed in the horizontal direction shown in FIG. The same classification is performed both when there is an edge in the direction and when there is an edge in the vertical direction shown in FIG. For this reason, since the direction in which the edge portion where the image quality is likely to deteriorate is correctly distinguished cannot be classified, the same prediction coefficient is used for the calculation even though there are different edge directions as described above. As a result, the image quality is deteriorated.
[0083]
In the present invention, the absolute values | Ch1-Cv1 |, | Cv1-Ct1 |, | Ct1-Ch1 of the differences obtained at the target pixel P1 by the processes of steps S5 and S6 (the flowcharts of FIGS. 8 and 12). In addition to |, the absolute values | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 | of the difference in phase of the pixel P2 present in the closest position in the right direction of the pixel of interest and the right of the pixel of interest By obtaining the absolute values | Ch3-Cv3 |, | Cv3-Ct3 |, | Ct3-Ch3 | of the phase difference of the pixel P2 present at the closest position in the direction, the horizontal direction in the case shown in FIG. And the vertical class code (code | Ch1-Cv1 |, code | Ch2-Cv2 |, code | Ch3-Cv3 |) are (1,1,1), as shown in FIG. Thus, when there is an edge in the vertical direction, the horizontal and vertical class codes (code | Ch1-Cv1 |, code | Ch2-Cv2 |, code | Ch3-Cv3 |) are (1, 0, 0). Therefore, even when classifying, it is possible to set the class by identifying the direction of the edge where image quality is likely to deteriorate, so it is possible to select the prediction coefficient more accurately for the pixel of interest Thus, image quality deterioration due to the conversion process can be suppressed.
[0084]
Also, as shown in FIG. 16, when there is a movement in the pixel of interest, that is, when there is a change in the time direction, the absolute difference between the chroma signal Ch1 in the horizontal direction and the chroma signal Ct1 in the time direction of the pixel of interest P1. When class classification processing is performed from the value, the difference exceeds the threshold th, and the presence of the motion of the pixel of interest P1 can be recognized. However, even when there is an edge in the vertical direction on the target pixel P1 as shown in FIG. 17, the absolute value | Ch1-Ct1 | exceeds the threshold th, and the pixel P1 has the target shown in FIG. Since the same class classification is performed when there is motion in a pixel and when there is an edge in the vertical direction shown in FIG. 17, the same prediction coefficient is obtained even though the pixel is in a different state. As a result, the image quality may be deteriorated.
[0085]
On the other hand, in the present invention, the absolute values | Ch1-Cv1 |, | Cv1-Ct1 |, | determined at the pixel of interest P1 by the processes of steps S5 and S6 (the flowcharts of FIGS. 8 and 12). In addition to Ct1-Ch1 |, the absolute values | Ch2-Cv2 |, | Cv2-Ct2 |, | Ct2-Ch2 | By obtaining the absolute values | Ch3-Cv3 |, | Cv3-Ct3 |, | Ct3-Ch3 | with respect to the pixel P2 having the opposite phase present at the closest position in the right direction, for example, as shown in FIG. When there is a motion, the class codes (code | Ct1-Ch1 |, code | Ct2-Ch2 |, code | Ct3-Ch3 |) in the time direction and the horizontal direction are (1, 1, 1). On the other hand, when there is an edge in the vertical direction as shown in FIG. 17, the class codes in the time direction and the horizontal direction (code | Ct1-Ch1 |, code | Ct2-Ch2 |, code | Ct3-Ch3 |) Is (1, 0, 0), and therefore can be identified even when classifying, so that it is possible to select a prediction coefficient more accurately and to suppress deterioration in image quality due to conversion processing. it can.
[0086]
For convenience of explanation, the description has been made using a 3-bit class code for the sake of convenience, but actually, as described above, a 9-bit class code (code | Cv1-Ct1 |, code | Ct1-Ch1 |, code | Ch2-Cv2 |, code | Cv2-Ct2 |, code | Ct2-Ch2 |, code | Ch3-Cv3 |, code | Cv3-Ct3 |, code | Ct3-Ch3 |).
[0087]
In the above description, the setting of the class code using the chroma signal C obtained by the simple Y / C separation process has been described. However, instead of the chroma signal C, the luminance signal obtained by the simple Y / C separation process. Y may be used.
[0088]
With the above processing, the class code can be determined by the chroma signal generated by the simple Y / C separation for each of the time direction, vertical direction, and horizontal direction of the pixel of interest and its neighboring pixels, and appropriate for each pixel. Since a conversion process from a composite signal to a component signal can be executed, deterioration in image quality can be suppressed.
[0089]
The selection of the class tap and the prediction tap is not limited to the examples shown in FIGS. 7 and 13, and any other pixel may be used as long as it is a pixel on a field close to the time direction, the vertical direction, and the horizontal direction. It may be a thing, and the number may be other numbers.
[0090]
Next, a learning device (prediction coefficient computing device) for determining the prediction coefficient stored in the coefficient memory 54 will be described with reference to FIG.
[0091]
The NTSC encoder 91 receives a component signal composed of a luminance signal Y and color difference signals RY and BY as a teacher image. The NTSC encoder 91 generates an NTSC composite video signal as a student image from the input component signal, and outputs it to the region extraction units 92 and 95.
[0092]
The area extraction unit 92 extracts pixels (class taps) necessary for class classification from the composite video signal converted into the input digital signal, and outputs the extracted pixels to the pattern detection unit 93. The pattern detection unit 93 detects the pattern of the composite video signal based on the input class tap. This pattern is a value obtained by Y / C separation processing in the horizontal direction, the vertical direction, and the time direction for each of the three pixels including the pixel of interest and the pixel of opposite phase to the pixel of interest closest to the left and right sides thereof. is there. The class code determination unit 94 determines a class code based on the pattern detected by the pattern detection unit 93 and outputs the class code to the normal equation generation unit 96. That is, the class code determination unit 94 performs Y / C separation in the vertical direction and the chroma signal Ch obtained by performing Y / C separation in the horizontal direction with respect to the signal of interest obtained by the pattern detection unit 93 and the pixels of opposite phase to the left and right. Chroma signal Cv, Y / C separated chroma signal Cv in the vertical direction and chroma signal Ct separated in Y / C in the time direction, and chroma signal Ct separated in Y / C in the time direction in Y / C separation in the horizontal direction The absolute value of the mutual difference is obtained from the combination of the chroma signals Ch, compared with a predetermined threshold value, a class code is determined based on the comparison result, and the class code is output to the normal equation generation unit 96.
[0093]
The region extraction unit 95 extracts a prediction tap from the composite video signal input from the NTSC encoder 91 and outputs the prediction tap to the normal equation generation unit 96. The region extraction unit 92, pattern detection unit 93, class code determination unit 94, and region extraction unit 95 described above are the region extraction unit 51, pattern detection unit 52, and class code determination unit 53 of the class classification adaptive processing circuit 22 in FIG. , And the region extraction unit 55 have basically the same configuration and function.
[0094]
The normal equation generation unit 96, for all classes input from the class code determination unit 94, for each class, the prediction tap of the student image input from the region extraction unit 95 and the luminance of the component signal as the teacher image A normal equation is generated from the signal Y and output to the coefficient determination unit 97. When the necessary number of normal equations are supplied from the normal equation generation unit 96, the coefficient determination unit 97 solves the normal equations using, for example, the least square method, and calculates the above-described prediction coefficients w1,. While calculating and determining, the prediction coefficient calculated | required by calculation is supplied to the memory 98, and is memorize | stored.
[0095]
Here, the normal equation will be described. In the above equation (1), the prediction coefficients w1,..., Wn are undetermined coefficients before learning. Learning is performed by inputting a plurality of teacher images for each class. When the number of types of teacher images is expressed as m, the following equation (20) is set from equation (1).
[0096]
yk = w1 * xk1 + w2 * xk2 ++ ... + wn * xkn (20)
[0097]
Here, k is k = 1, 2,..., M. When m> n, the prediction coefficients w1,..., wn are not uniquely determined. Therefore, the element ek of the error vector e is defined by the following equation (21), and the prediction coefficient is determined so that the error vector e defined by the equation (22) is minimized. That is, for example, the prediction coefficient is uniquely determined by a so-called least square method.
[0098]
ek = yk- {w1 * xk1 + w2 * xk2 ++ ... + wn * xkn} (21)
[0099]
[Expression 1]

[0100]
Each prediction coefficient wi that minimizes e ^ 2 (e squared) of Expression (21) is obtained by partial differentiation of e ^ 2 with the prediction coefficient w i (i = 1, 2,...) (23) The partial differential value is calculated to be 0 for each value of i.
[0101]
[Expression 2]

[0102]
A specific procedure for determining each prediction coefficient w i from Equation (23) will be described. When defining Xji and Yi as in the equations (24) and (25), the equation (23) is transformed into the determinant of the equation (26).
[0103]
[Equation 3]

[Expression 4]

[Equation 5]

[0104]
This equation (26) is generally called a normal equation. Here, Xji (j, i = 1, 2,... N) and Yi (i = 1, 2,... N) are calculated based on the teacher image and the student image. That is, the normal equation generation unit 96 calculates the values of Xji and Yi and determines the equation (26) consisting of the normal equation. Further, the coefficient determination unit 97 determines each prediction coefficient wi by solving the equation (26).
[0105]
Next, with reference to the flowchart of FIG. 19, the process in which the learning apparatus of FIG. 18 learns the prediction coefficient will be described.
[0106]
In step S 61, a component signal as a teacher image is input to the NTSC encoder 91 and the normal equation generation unit 96. In step S <b> 62, the NTSC encoder 91 generates a student image composed of an NTSC composite signal from the input component signal and outputs the student image to the region extraction units 92 and 95.
[0107]
In step S <b> 63, the region extraction unit 92 executes region extraction processing, extracts class taps, and outputs them to the pattern detection unit 93. This process is the same as the process in step S4 in the flowchart of FIG.
[0108]
In step S <b> 64, the pattern detection unit 93 executes pattern detection processing, detects a pattern for determining a class code from the class tap, and outputs the pattern to the class code determination unit 94. This process is the same as the process of step S5 in the flowchart of FIG. 6, and is similar to the process described with reference to the flowchart of FIG.
[0109]
In step S <b> 65, the class code determination unit 94 executes class code determination processing based on the pattern detected by the pattern detection unit 93, and outputs it to the normal equation generation unit 96. This process is the same as the process of step S6 in the flowchart of FIG. 6, and is the same as the process described with reference to the flowchart of FIG.
[0110]
In step S <b> 66, the region extraction unit 95 extracts the prediction tap of the student image and outputs it to the normal equation generation unit 96.
[0111]
In step S67, the normal equation generation unit 96 uses the class code input from the class code determination unit 94, the prediction tap input from the region extraction unit 95, and the component signal input as a teacher image to calculate the above equation (26). ) And is output to the coefficient determination unit 97 together with the class code input from the class code determination unit 94.
[0112]
In step S68, the coefficient determining unit 97 determines the prediction coefficient by solving the normal equation composed of the above-described equation (26), and stores it in the memory 98 in association with the class code.
[0113]
In step S69, it is determined whether or not processing has been performed for all pixels. If it is determined that processing has not been performed for all pixels, the processing returns to step S63. That is, the processes in steps S63 to S69 are repeated until the processes for all the pixels are completed. If it is determined in step S69 that all the pixels have been processed, the processing ends.
[0114]
With the above processing, the class code can be determined by the chroma signal generated by the simple Y / C separation for each of the time direction, the vertical direction, and the horizontal direction of the target pixel and its neighboring pixels. Therefore, since an appropriate prediction coefficient can be set, conversion processing from an appropriate composite signal to a component signal can be executed for each pixel, and deterioration in image quality can be suppressed.
[0115]
Further, in the above, the absolute values | Cv1-Ct1 |, | Ct1-Ch1 |, | Ch2-Cv2 of the differences of the chroma signals C having different directions when performing simple Y / C separation in the class code determination units 53 and 94 |, | Cv2-Ct2 |, | Ct2-Ch2 |, | Ch3-Cv3 |, | Cv3-Ct3 |, and | Ct3-Ch3 | are set to a binary class code by comparing with a predetermined threshold th. In the example described above, for example, a ternary class code is determined based on the magnitude relationship between the difference itself and two threshold values th and −th (| th | = | −th |) having the same absolute value. You may do it. FIG. 20 shows the difference of the chroma signal obtained by Y / C separation from different directions, and the class code determination unit 53 in the case of determining the class code from the magnitude relationship with the two threshold values th and -th. The configuration is shown.
[0116]
When a chroma signal Y / C separated from different directions is represented by Cm, Cn ((Cm, Cn) is either (Cv, Ch), (Ch, Ct) (Ct, Cv)) When the signals Cm and Cn are input to the class code determination unit 53, the subtraction unit 111 calculates the difference (Cm−Cn) and outputs the difference (Cm−Cn) to the comparator 112. The comparator 112 compares the input difference (Cm−Cn) with the threshold values th and −th. In this case, when the difference (Cm−Cn) is equal to or smaller than the threshold −th, the comparator 112 sets the class code to 0 and the difference (Cm−Cn) is larger than the threshold −th and equal to or smaller than the threshold th. In this case, the class code is set to 1, and the class code is set to 2 when the difference (Cm−Cn) is larger than the threshold th. By setting the class code in this way, 3 ^ 9 (3 to the 9th power) class codes can be set, and the direction of change of the chroma signal can be reflected in the class code. In addition, it is possible to perform classification processing in consideration of the direction of the change.
[0117]
In the above example, the component video signal is generated from the composite video signal. For example, as shown in FIG. 21, the class classification adaptive processing circuit 22 and the matrix circuit 6 of FIG. The class classification adaptive processing circuit 121 may be configured to directly generate the primary color RGB signal from the composite video signal output from the VIF circuit 3. Also in this case, the configuration of the class classification adaptive processing circuit 121 is the same as that of the class classification adaptive processing circuit 22, but the student image in the learning process for determining the prediction coefficient stored in the coefficient memory 54 is a composite signal. However, the teacher image becomes the primary color RGB signal, and the determined prediction coefficient also corresponds to the primary color RGB signal.
[0118]
Furthermore, in the above, the example using the NTSC composite signal has been described. However, for example, when converting from a PAL (Phase Alternating Line) composite signal to a component signal, the same configuration is used. Can do.
[0119]
According to the above, Y / C separation, color demodulation, matrix processing, and image information conversion processing can be performed collectively by class classification adaptive processing, so the apparatus scale can be reduced and costs can be reduced. In addition, it is possible to reduce image quality degradation in the conversion from composite signal to component signal, particularly image quality degradation due to the inability to classify class codes that correctly distinguish the edge directions that are likely to cause image quality degradation. .
[0120]
The series of processes described above can be executed by hardware, but can also be executed by software. When a series of processes is executed by software, a program constituting the software may execute various functions by installing a computer incorporated in dedicated hardware or various programs. For example, it is installed from a recording medium in a general-purpose personal computer or the like.
[0121]
FIG. 22 shows a configuration of an embodiment of a personal computer when the television receiver or the learning device is realized by software. The CPU 201 of the personal computer controls the entire operation of the personal computer. Further, when a command is input from the input unit 206 such as a keyboard or a mouse from the user via the bus 204 and the input / output interface 205, the CPU 201 is stored in a ROM (Read Only Memory) 202 correspondingly. Run the program. Alternatively, the CPU 201 reads a program read from the magnetic disk 211, the optical disk 212, the magneto-optical disk 213, or the semiconductor memory 214 connected to the drive 210 and installed in the storage unit 208 into a RAM (Random Access Memory) 203. To load and execute. Thereby, the functions of the above-described image processing apparatus are realized by software. Further, the CPU 201 controls the communication unit 209 to communicate with the outside and exchange data.
[0122]
As shown in FIG. 22, the recording medium on which the program is recorded is distributed to provide the program to the user separately from the computer, and the magnetic disk 211 (including the flexible disk) on which the program is recorded, By a package medium comprising an optical disk 212 (including compact disc-read only memory (CD-ROM), DVD (digital versatile disc)), a magneto-optical disk 213 (including MD (mini-disc)), or a semiconductor memory 214 In addition to being configured, it is configured by a ROM 202 on which a program is recorded, a hard disk included in the storage unit 208, and the like provided to the user in a state of being incorporated in a computer in advance.
[0123]
In this specification, the step of describing the program recorded on the recording medium is not limited to the processing performed in time series in the order described, but of course, it is not necessarily performed in time series. Or the process performed separately is included.
[0124]
【The invention's effect】
  According to the image information conversion apparatus and method and the program of the present invention, the pixel signal of the pixel of interest, And pixel signals around the pixel of interestFrom the input composite signalSeparated into luminance signal and chroma signal for each given spatio-temporal direction,For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, the difference between luminance signals separated in different spatiotemporal directions or chroma signals separated in different spatiotemporal directions is calculated and calculated resultEach with a predetermined threshold, and the comparison resultConsist ofGenerate pattern, patternEveryClassify the pixel of interest into one of multiple classesFor converting the pixel of interestExtract pixel signals of multiple pixels, pixel signals of multiple pixels and classified classesIs set toWith coefficientSum of productsThe pixel signal of the pixel of interest is converted by calculation.
[0126]
  According to the coefficient calculation apparatus, method, and program of the present invention, an input composite signal is generated from an input image signal, and a pixel signal of a pixel of interest, And pixel signals around the pixel of interestFrom the input composite signalSeparated into luminance signal and chroma signal for each given spatio-temporal direction,For each pixel signal of the pixel of interest and each pixel signal around the pixel of interest, the difference between luminance signals separated in different spatiotemporal directions or chroma signals separated in different spatiotemporal directions is calculated and calculated resultEach with a predetermined threshold, and the comparison resultConsist ofGenerate pattern, patternEveryClassify the pixel of interest into one of multiple classesFor converting the pixel of interestExtract pixel signals of multiple pixels and input pixel signals of multiple pixels and image signalsFrom the normal equation using and,classEveryThe coefficient was calculated.
[0127]
In any case, as a result, the apparatus scale can be reduced, the cost can be reduced, and the image quality deterioration in the conversion from the composite signal to the component signal, in particular, the error in the class code classification process. It is possible to reduce image quality degradation that is the cause.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration example of a conventional television receiver.
FIG. 2 is a block diagram illustrating a configuration example of a conventional television receiver.
FIG. 3 is a block diagram showing a configuration of an embodiment of a television receiver to which the present invention is applied.
4 is a block diagram showing a configuration example of a class classification adaptive processing circuit in FIG. 1. FIG.
5 is a block diagram illustrating a configuration example of a class code determination unit in FIG. 4. FIG.
6 is a flowchart illustrating display processing of the television receiver in FIG. 3. FIG.
FIG. 7 is a diagram illustrating an example of a class tap.
FIG. 8 is a flowchart illustrating pattern detection processing.
FIG. 9 is a diagram illustrating a simple Y / C separation process.
FIG. 10 is a diagram illustrating a simple Y / C separation process.
FIG. 11 is a diagram illustrating a simple Y / C separation process.
FIG. 12 is a flowchart illustrating class code determination processing.
FIG. 13 is a diagram illustrating an example of a prediction tap.
FIG. 14 is a diagram illustrating a case where there is a horizontal edge on a pixel of interest.
FIG. 15 is a diagram illustrating a case where there is an edge in the vertical direction on a pixel of interest.
FIG. 16 is a diagram for describing a case where a target pixel has a motion.
FIG. 17 is a diagram illustrating a case where there is an edge in the vertical direction on a pixel of interest.
FIG. 18 is a block diagram illustrating a learning device.
FIG. 19 is a flowchart illustrating a learning process of the learning device.
FIG. 20 is a block diagram illustrating another configuration example of the class code determination unit.
FIG. 21 is a block diagram illustrating another configuration example of the television receiver.
FIG. 22 is a diagram illustrating a medium.
[Explanation of symbols]
22 class classification adaptive processing circuit, 51 region extraction unit, 52 pattern detection unit, 53 class code determination unit, 54 coefficient memory, 55 region extraction unit, 56 prediction operation unit, 71 subtractor, 72 ABS circuit, 73 comparator, 91 NTSC encoder, 92 region extraction unit, 93 pattern detection unit, 94 class code determination unit, 95 region extraction unit, 96 normal equation generation unit, 97 coefficient determination unit, 98 memory, 111 subtractor, 112 comparator, 121 class classification adaptation Processing circuit

Claims (16)

Separating means for separating a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The luminance signal separated in different spatiotemporal directions or the chroma separated in different spatiotemporal directions by the separating means for each pixel signal of the pixel of interest and pixel signals around the pixel of interest. Difference calculating means for calculating a difference between signals;
Comparison means for comparing the calculation results of the difference calculation means with a predetermined threshold value;
Pattern generation means for generating a pattern composed of a comparison result by the comparison means;
Class classification means for classifying the pixel of interest into one of a plurality of classes for each pattern;
Extraction means for extracting pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
Pixel-of- interest signal conversion means for converting the pixel signal of the pixel of interest by a product-sum operation of the pixel signals of the plurality of pixels and the coefficients set in the class classified by the class classification means. An image information conversion device characterized by the above. The separation means includes three pixels obtained by adding the pixel of interest to two pixels closest in the left-right direction in the spatio-temporal direction, the pixel having the phase opposite to the pixel of interest, and the time Three pixels obtained by adding the pixel of interest to the two pixels closest in the vertical direction in the spatial direction , and the pixel of interest in one pixel closest in the time direction in the spatio-temporal direction and in the opposite phase to the pixel of interest The pixel signal of the pixel of interest is separated into a luminance signal and a chroma signal in the horizontal direction, the vertical direction, and the time direction, respectively, and the pixel signal of the pixel of interest 3 pixels obtained by adding the peripheral pixels of the pixel of interest to the two pixels closest in the left-right direction in the spatio-temporal direction, the reverse phase of the peripheral pixels of the pixel of interest, , the time space direction Three of pixels plus peripheral pixels of the target pixel in the vertical direction to the nearest two pixels, and said around the target pixel and the pixel in opposite phase, and the last in the time direction of the space-time direction 1 Using the pixel signals of two pixels obtained by adding pixels around the pixel of interest to the pixel, the luminance of the pixel signals around the pixel of interest in the left-right direction, the up-down direction, and the time direction, respectively, The image information conversion apparatus according to claim 1, wherein the image information conversion apparatus is separated into a signal and the chroma signal . 2. The image according to claim 1, wherein pixels around the pixel of interest are pixels that have an opposite phase to the pixel of interest and are closest to the left-right direction , the up-down direction, or the time direction. Information conversion device. The difference calculation means includes a difference between the luminance signal separated in the left-right direction and the luminance signal separated in the up-down direction for each pixel signal of the pixel of interest and pixel signals around the pixel of interest, the vertical direction The difference between the luminance signal separated in the time direction and the luminance signal separated in the time direction, the difference between the luminance signal separated in the time direction and the luminance signal separated in the left-right direction, or the chroma signal separated in the left-right direction And the difference between the chroma signals separated in the vertical direction, the difference between the chroma signal separated in the vertical direction and the chroma signal separated in the time direction, and the chroma signal separated in the time direction and the horizontal signal The image information conversion apparatus according to claim 1, wherein a difference between luminance chromas is calculated . The chroma signals separated in the different spatio-temporal directions or the chromas separated in the different spatio-temporal directions for each pixel signal of the pixel of interest and the pixel signals around the pixel of interest by the difference calculation means. An absolute value calculating means for calculating the absolute value of the difference between the signals ,
The comparison means, as a calculation result of the difference calculation means, between the luminance signals separated in the different spatiotemporal directions or different for each pixel signal of the pixel of interest and each pixel signal around the pixel of interest The image information conversion apparatus according to claim 1, wherein absolute values of differences between the chroma signals separated in a spatio-temporal direction are respectively compared with a predetermined threshold value. The pixel-of-interest signal conversion means converts the pixel signal of the pixel of interest into a luminance signal by a product-sum operation of the pixel signals of the plurality of pixels and the coefficient set for each class classified by the class classification means . Or it converts into a chroma signal . The image information converter of Claim 1 characterized by the above-mentioned. The pixel-of-interest signal conversion unit converts the pixel signal of the pixel of interest into a component signal by performing a product-sum operation on the pixel signals of the plurality of pixels and the coefficient set for each class classified by the class classification unit The image information conversion device according to claim 1, wherein conversion is performed. Display means for displaying the pixel signal of the pixel of interest converted into the component signal is further provided.
  The image information conversion apparatus according to claim 1. A separation step of separating a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or separated in different spatiotemporal directions by the processing of the separation step. A difference calculating step for calculating a difference between the chroma signals;
A comparison step of comparing each of the calculation results in the difference calculation step with a predetermined threshold;
A pattern generation step for generating a pattern composed of a comparison result in the process of the comparison step;
A class classification step for classifying the target pixel into one of a plurality of classes for each pattern;
An extraction step of extracting pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
A pixel-of- interest signal conversion step of converting a pixel signal of the pixel of interest by a product-sum operation of the pixel signals of the plurality of pixels and the coefficients set in the class classified by the processing of the class classification step; An image information conversion method comprising: A separation control step for controlling separation of a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or in different spatiotemporal directions by the process of the separation control step. A difference calculation control step for controlling calculation of a difference between the chroma signals;
A comparison control step for controlling the comparison of each calculation result in the difference calculation control step with a predetermined threshold;
A pattern generation control step for controlling generation of a pattern composed of a comparison result in the process of the comparison control step;
A class classification control step for controlling the classification of the pixel of interest into one of a plurality of classes for each pattern;
An extraction control step for controlling extraction of pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
The pixel of interest that controls the conversion of the pixel signal of the pixel of interest by the product-sum operation of the pixel signals of the plurality of pixels and the coefficient set in the class whose classification is controlled in the processing of the class classification control step A recording medium on which a computer-readable program is recorded, comprising: a signal conversion control step. A separation control step for controlling separation of a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or in different spatiotemporal directions by the process of the separation control step. A difference calculation control step for controlling calculation of a difference between the chroma signals;
A comparison control step for controlling the comparison of each calculation result in the difference calculation control step with a predetermined threshold;
A pattern generation control step for controlling generation of a pattern composed of a comparison result in the process of the comparison control step;
A class classification control step for controlling the classification of the pixel of interest into one of a plurality of classes for each pattern;
An extraction control step for controlling extraction of pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
The pixel of interest for controlling the conversion of the pixel signal of the pixel of interest by the product-sum operation of the pixel signals of the plurality of pixels and the coefficient set for the class whose classification is controlled in the processing of the class classification control step A program that causes a computer to execute a signal conversion control step. Composite signal generation means for generating an input composite signal from the input image signal;
Separating means for separating a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The luminance signal separated in different spatiotemporal directions or the chroma separated in different spatiotemporal directions by the separating means for each pixel signal of the pixel of interest and pixel signals around the pixel of interest. Difference calculating means for calculating a difference between signals;
Comparison means for comparing the calculation results of the difference calculation means with a predetermined threshold value;
Pattern generation means for generating a pattern composed of a comparison result by the comparison means;
Class classification means for classifying the pixel of interest into one of a plurality of classes for each pattern;
Extracting means for extracting pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
A coefficient calculating apparatus comprising: a calculating unit that calculates a coefficient for each class from a normal equation using pixel signals of the plurality of pixels and the input image signal. A storage device that stores the coefficient calculated by the coefficient calculation device according to claim 12 . A composite signal generation step for generating an input composite signal from the input image signal;
A separation step of separating a pixel signal of the pixel of interest and a pixel signal around the pixel of interest from an input composite signal into a luminance signal and a chroma signal for each of a predetermined spatiotemporal direction ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or separated in different spatiotemporal directions by the processing of the separation step. A difference calculating step for calculating a difference between the chroma signals;
A comparison step of comparing each of the calculation results in the difference calculation step with a predetermined threshold;
A pattern generation step for generating a pattern composed of a comparison result in the process of the comparison step;
A class classification step for classifying the target pixel into one of a plurality of classes for each pattern;
An extraction step of extracting pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
A coefficient calculation method comprising: a calculation step of calculating a coefficient for each class from a normal equation using pixel signals of the plurality of pixels and the input image signal. A composite signal generation control step for controlling the generation of an input composite signal from the input image signal;
A separation control step for controlling separation into a luminance signal and a chroma signal for each of a predetermined spatio-temporal direction from an input composite signal of a pixel signal of the pixel of interest and a pixel signal around the pixel of interest ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or in different spatiotemporal directions by the process of the separation control step. A difference calculation control step for controlling calculation of a difference between the chroma signals;
A comparison control step for controlling the comparison of each calculation result in the difference calculation control step with a predetermined threshold;
A pattern generation control step for controlling generation of a pattern composed of a comparison result in the process of the comparison control step;
A class classification control step for controlling the classification of the pixel of interest into one of a plurality of classes for each pattern;
An extraction control step for controlling extraction of pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
A computer-readable program comprising: a calculation control step for controlling calculation of a coefficient for each class from a normal equation using pixel signals of the plurality of pixels and the input image signal. Recorded recording medium. A composite signal generation control step for controlling the generation of an input composite signal from the input image signal;
A separation control step for controlling separation into a luminance signal and a chroma signal for each of a predetermined spatio-temporal direction from an input composite signal of a pixel signal of the pixel of interest and a pixel signal around the pixel of interest ;
The pixel signals of the pixel of interest and the pixel signals around the pixel of interest are separated between the luminance signals separated in different spatiotemporal directions or in different spatiotemporal directions by the process of the separation control step. A difference calculation control step for controlling calculation of a difference between the chroma signals;
A comparison control step for controlling the comparison of each calculation result in the difference calculation control step with a predetermined threshold;
A pattern generation control step for controlling generation of a pattern composed of a comparison result in the process of the comparison control step;
A class classification control step for controlling the classification of the pixel of interest into one of a plurality of classes for each pattern;
An extraction control step for controlling extraction of pixel signals of a plurality of pixels for converting the pixel of interest from the input composite signal;
A program for causing a computer to execute a calculation control step of controlling calculation of a coefficient for each class from a normal equation using pixel signals of the plurality of pixels and the input image signal.

Images