JP2004056414A

JP2004056414A - Electronic watermark information embedding apparatus and embedding method

Info

Publication number: JP2004056414A
Application number: JP2002210474A
Authority: JP
Inventors: Katsuyoshi Nishitani; 西谷　勝義; Takayuki Sugawara; 菅原　隆幸
Original assignee: Victor Company of Japan Ltd
Current assignee: Victor Company of Japan Ltd
Priority date: 2002-07-19
Filing date: 2002-07-19
Publication date: 2004-02-19
Anticipated expiration: 2022-07-19
Also published as: JP3829770B2

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electronic watermark information embedding apparatus capable of simply embedding electronic watermark information into an image, and to provide an embedding method therefor. <P>SOLUTION: An average value of the luminance values of image signals of even number lines is calculated in odd number lines and the even number lines of a particular area of the image signals designated by an area designation unit 1 to compare the average value with a luminance value of each pixel of the odd number lines in the particular area. When the average value is smaller, a counter CNTposi is incremented by 1, and when the average value is greater, a counter CNTnega is incremented by 1. The luminance values of the image signals of the odd number lines and the even number lines in the particular area are corrected in a reverse direction to each other so that a relation of CNTposi>CNTnega holds when a bit of encryption information to be embedded from an encryption generator 2 is '1' and a relation of CNTposi<CNTnega holds when the bit is '0'. <P>COPYRIGHT: (C)2004,JPO

Description

【０００１】
【発明の属する技術分野】
本発明は電子透かし情報埋め込み装置及び埋め込み方法に係り、特に著作権の所有権を証明し、著作権侵害を識別し、隠されたメッセージを送信するために、画像、音声、マルチメディアデータに電子透かし（ウォーターマーク）情報を埋め込む電子透かし情報埋め込み装置及び埋め込み方法に関する。
【０００２】
【従来の技術】
電子透かし信号は画像や音声のマルチメディアデータに何らかの情報を埋め込み、隠し持たせる技術である。埋め込む方式は様々なものがある。従来より知られている電子透かし情報埋め込み方法として、例えば、情報をＭＰＥＧ方式により圧縮符号化された符号、特にＤＣＴ係数や、動きベクトル、量子化特性の変更に基づく情報埋め込み手法が提案されている（小川宏他２名、”ＤＣＴを用いたデジタル動画像における著作権情報埋め込み方法”、ＳＣＩＳ’９７−３１Ｇ）。
【０００３】
また、他の従来の電子透かし情報埋め込み方法として、直接拡散方式に従い、ＰＮ系列で画像信号を拡散し、画像に署名情報を合成する方法も知られている（大西淳児他２名、”ＰＮ系列による画像への透かし署名法”、ＳＣＩＳ’９７＿２６Ｂ）。この従来方法では、署名を含んだ画像信号を逆拡散すると、署名情報は画像信号全体に拡散し、拡散された信号は非常に弱く、画像信号に対して大きなノイズとはならず、署名情報を含んだ画像信号は見かけ上は原画像と同じである。署名情報を確認するには、拡散符号で画像信号を拡散することにより、署名情報の信号を検出する。
【０００４】
【発明が解決しようとする課題】
しかるに、上記の従来の電子透かし情報埋め込み方法では、透かし情報を画像に埋め込んで送る場合、画像の変換処理に時間、コストがかかるという問題がある。また、ＤＣＴ等の直交変換を用いて算出された変換係数に対して、透かし情報を埋め込むと、簡単な低域フィルタでも透かし情報が欠損する可能性があるという問題もある。
【０００５】
本発明は以上の点に鑑みなされたもので、簡単に電子透かし情報を画像に埋め込み得る電子透かし情報埋め込み装置及び埋め込み方法を提供することを目的とする。
【０００６】
また、本発明の他の目的は、フィルタに対して耐性の強い電子透かし情報の記録再生を行い得る電子透かし情報埋め込み装置及び埋め込み方法を提供することにある。
【０００７】
【課題を解決するための手段】
上記の目的を達成するため、本発明の電子透かし情報埋め込み装置は、入力された画像信号中の電子透かし情報を埋め込む特定領域を指定する領域指定手段と、電子透かし情報を暗号化して暗号情報として出力する暗号発生手段と、入力された画像信号中の特定領域内の第１の領域の全画素の輝度値又は色差値の平均値と特定領域内の第１の領域以外の第２の領域の各画素の輝度値又は色差値とを大小比較して、第２の領域の全画素のうち、輝度値又は色差値が平均値よりも大きい画素数を示す第１のカウンタ値と、輝度値又は色差値が平均値よりも小さい画素数を示す第２のカウンタ値を求め、その第１及び第２のカウンタ値の大小関係を、暗号発生手段からの暗号情報の第１の値又は第２の値に応じて予め設定した関係とするように、第１の領域と第２の領域の各画素の輝度値又は色差値を相反する方向に変換し、電子透かし情報が埋め込まれた画像信号として出力する混入手段とを有する構成としたものである。
【０００８】
この発明では、電子透かし情報を暗号化した暗号情報に応じて、入力画像信号中の特定領域の第１の領域と第２の領域の各画素の輝度値又は色差値を相反する方向に変換し、電子透かし情報が埋め込まれた画像信号として出力するようにしたため、第１及び第２の領域として隣接する画像ラインを指定することにより、隣接する画像ライン間の相関性が非常に高いという性質を利用した電子透かし情報の埋め込みができる。
【０００９】
ここで、上記の混入手段は、入力された画像信号中の特定領域内の第１の領域の画素の輝度値又は色差値の平均値を算出する平均値算出手段と、平均値と特定領域内の第１の領域以外の第２の領域の各画素とを大小比較し、第２の領域の全画素のうち、輝度値又は色差値が平均値よりも大きい画素数を示す第１のカウンタ値と、輝度値又は色差値が平均値よりも小さい画素数を示す第２のカウンタ値とを算出するカウンタ値算出手段と、第１及び第２のカウンタ値を大小比較するカウンタ値比較手段と、暗号発生手段からの暗号情報の値が第１の値のときには、カウンタ値比較手段により第１のカウンタ値が第２のカウンタ値よりも大なる比較結果が得られ、暗号発生手段からの暗号情報の値が第２の値のときには、カウンタ値比較手段により第１のカウンタ値が第２のカウンタ値よりも小なる比較結果が得られるように、第１の領域と第２の領域の全画素の輝度値又は色差値を相反する方向に変換する変換手段とよりなり、変換手段により第１の領域と第２の領域の各画素の輝度値又は色差値が暗号情報の値に応じて変換された信号を、電子透かし情報が埋め込まれた画像信号として出力することを特徴とする。
【００１０】
また、上記の目的を達成するため、本発明の埋め込み方法は、入力された画像信号中の電子透かし情報を埋め込む特定領域を指定する第１のステップと、電子透かし情報を暗号化して暗号情報として出力する第２のステップと、入力された画像信号中の特定領域内の第１の領域の全画素の輝度値又は色差値の平均値と特定領域内の第１の領域以外の第２の領域の各画素の輝度値又は色差値とを大小比較して、第２の領域の全画素のうち、輝度値又は色差値が平均値よりも大きい画素数を示す第１のカウンタ値と、輝度値又は色差値が平均値よりも小さい画素数を示す第２のカウンタ値を求める第３のステップと、第１及び第２のカウンタ値の大小関係を、暗号情報の第１の値又は第２の値に応じて予め設定した関係とするように、第１の領域と第２の領域の各画素の輝度値又は色差値を相反する方向に変換し、電子透かし情報が埋め込まれた画像信号として出力する第４のステップとを含むことを特徴とする。
【００１１】
この発明では、電子透かし情報を暗号化した暗号情報に応じて、入力画像信号中の特定領域の第１の領域と第２の領域の各画素の輝度値又は色差値を相反する方向に変換し、電子透かし情報が埋め込まれた画像信号として出力するようにしたため、第１及び第２の領域として隣接する画像ラインを指定することにより、隣接する画像ライン間の相関性が非常に高いという性質を利用した電子透かし情報の埋め込みができる。
【００１２】
【発明の実施の形態】
次に、本発明の実施の形態について図面と共に説明する。図１は本発明になる電子透かし記録再生装置の記録部の一実施の形態のブロック図を示す。同図において、画像信号は領域指定器１に供給され、ここで電子透かし情報を埋め込まれる領域が決定される。
【００１３】
一方、暗号発生器２では、画像信号に埋め込まれる電子透かし情報を暗号化する。暗号化の間単な一例としては、ＡＳＣＩＩ文字を送る場合、文字は１バイト（８ビット）で表されるので、８個のビットを１ビットずつ順に送るものとすればよいし、また、後述する図２の電子透かし情報再生部の暗号復号器６との整合性がとれれば、ハフマンテーブルのようなテーブルを用いて符号化したものを１ビットずつ送ってもよい。
【００１４】
暗号混入器３は、領域指定器１により指定された画像信号の特定領域に、暗号発生器２から送出される暗号情報に基づいて、画像信号の輝度値を変換する。ここでは、図３に点線で示した、画像信号の上端２ラインの領域１０を特定領域とし（図３中、白丸が奇数ラインの画素、黒丸が偶数ラインの画素を示す）、画像信号の１ライン当りの有効画素数をＮ画素とする。
【００１５】
特定領域１０の奇数ラインと偶数ラインにおいて、まず、偶数ラインの画像信号の輝度値の平均値Ｐ_ＡＶＥ（ＥＶＥＮ）を次式により算出する。
【００１６】
【数１】

ただし、（１）式中、Ｐｉ（ＥＶＥＮ）は、特定領域１０の偶数ラインの画像信号の輝度値を示す。
【００１７】
次に算出された偶数ラインの画像信号の輝度値の平均値Ｐ_ＡＶＥ（ＥＶＥＮ）と、特定領域１０の奇数ラインの画像信号の輝度値Ｐｉ（ＯＤＤ）とを比較し、次式の不等式
Ｐｉ（ＯＤＤ）−Ｐ_ＡＶＥ（ＥＶＥＮ）＞０　　（ｉ＝１，２，．．，Ｎ）　　（２）
が成立するときには、正のカウンタＣＮＴｐｏｓｉを１インクリメントし、逆に
Ｐｉ（ＯＤＤ）−Ｐ_ＡＶＥ（ＥＶＥＮ）＜０　　（ｉ＝１，２，．．，Ｎ）　　（３）
が成立するときには、負のカウンタＣＮＴｎｅｇａを１インクリメントする。この比較を特定領域内の奇数ラインのすべての画素の輝度値に対して行う。
【００１８】
ここで、予め電子透かし情報記録部と、電子透かし情報再生部での暗号情報の意味付けを設定しており、ここでは埋め込む暗号情報のビットが”１”である場合、
ＣＮＴｐｏｓｉ＞ＣＮＴｎｅｇａ　　　　　　　　　　　　　　　　（４）
となり、埋め込む暗号情報のビットが”０”である場合は、
ＣＮＴｐｏｓｉ＜ＣＮＴｎｅｇａ　　　　　　　　　　　　　　　　（５）
となるように、奇数ラインの画像信号の輝度値及び偶数ラインの画像信号の輝度値を、奇数ラインの画像信号の輝度値を修正する方向とは相反する方向に修正する。
【００１９】
（４）式を満足するためには（暗号情報のビット”１”を埋め込むためには）、奇数ラインの画像信号の輝度値に対して、すべて１ずつ輝度レベルを上げ、再度偶数ラインの画像信号の輝度値の平均値Ｐ_ＡＶＥ（ＥＶＥＮ）と比較を行い、もし（４）式を満足すれば、処理を終了する。しかし、依然として（４）式を満足しない場合は、再度奇数ラインの画像信号の輝度値に対して、すべて１ずつ輝度レベルを上げ、再び偶数ラインの画像信号の輝度値の平均値Ｐ_ＡＶＥ（ＥＶＥＮ）と比較を行うことを、（４）式を満足するようになるまで繰り返す。
【００２０】
このとき、輝度値のレベルが２以上上がる画素においては、インクリメントする動作の偶数番目は、偶数ライン側の画像信号の輝度値を、奇数ライン側の画像信号の輝度値を修正する方向とは相反する方向に修正することで、相対的に、偶数ラインの画像信号の輝度値から見て奇数ラインの画像信号の輝度値を２回インクリメントしたと同等になるような処理を施す。
【００２１】
同様にして、暗号情報のビット”０”を埋め込む場合、奇数ラインの画像信号の輝度値に対して、すべて１ずつ輝度レベルを下げる修正を行ってから、再度偶数ラインの画像信号の輝度値の平均値Ｐ_ＡＶＥ（ＥＶＥＮ）と比較を行うことを、（５）式を満足するまで繰り返す。
【００２２】
このとき、輝度値のレベルが２以上下がる画素においては、デクリメントする動作の偶数番目は、偶数ライン側の画像信号の輝度値を、奇数ライン側の画像信号の輝度値を修正する方向とは相反する方向に修正することで、相対的に、偶数ラインの画像信号の輝度値から見て奇数ラインの画像信号の輝度値を２回デクリメントしたと同等になるような処理を施す。
【００２３】
以上の手法では、隣接する画像ライン間の相関が非常に高いことから、輝度レベルの修正も比較的小さな値で済むため、視覚特性上の画質劣化も認知しにくいレベルに抑えられ、簡単に暗号情報の埋め込みが実現できる。こうして、暗号情報のビットに基づいて、特定領域の奇数ラインの各画素の輝度値を修正された画像信号が、電子透かし埋め込み画像信号として暗号混入器３から出力され、図示しない公知の記録手段により記録媒体に記録される。
【００２４】
ここで、片側の特定領域の画素値だけを修正する場合は、図４で表される。すなわち、図４は奇数ラインのみの各画素の輝度値を修正されると仮定した場合の様子を示しており、偶数ラインのデータを点線ですべて基準の値として仮定し、奇数ラインのデータのみを例えば＋２、若しくは−２する例を示す。この図４における画素修正値とは、特定の１画素の輝度値の場合もあるし、ラインの輝度平均値の場合もあり得ることを概念的に示す。
【００２５】
これに対し、本発明では特定領域の片側のラインの輝度値だけを修正するのではなく、図５に示すように、特定領域の奇数ラインと偶数ラインの両方の輝度値を相反する方向へ修正する。すなわち、偶数ラインと奇数ラインの輝度値差を相対的に見れば図４と同じであるが、双方のラインを修正することで奇数及び偶数のうちの片側のライン（図４では奇数ライン）だけを修正するときに、変化させる値の半分の値で済ませることも可能となる。
【００２６】
この奇数ラインと偶数ラインの輝度値を相反する方向へ修正するという方法は、片側のラインだけ使用すると仮定した場合の輝度値の変化量を、少なくとも２以上変化させるときに使用する。すなわち、図４のような奇数ラインのみの各画素の輝度値を修正する場合に比較して、奇数ラインと偶数ラインの各画素の輝度値を相反する方向へ修正する場合は、図５に示すようになる。
【００２７】
図５では、偶数ラインのデータと奇数ラインのデータの相対的な差を同じにして、その差の中央を点線で基準の値として仮定し、奇数ラインのデータを例えば＋１、若しくは−１し、偶数ラインのデータを相反する方向に、例えば−１、若しくは＋１する例を示したものである。この図５における画素修正値とは、特定の１画素の輝度値の場合もあるし、ラインの輝度平均値の場合もあることを概念的に示す。
【００２８】
次に、再生部の構成及び動作について説明する。図２は電子透かし記録再生装置の再生部の一実施の形態のブロック図を示す。同図において、記録媒体から公知の再生手段により再生された電子透かし埋め込み画像信号は、領域指定器４に入力され、ここで電子透かし情報記録部の領域指定器１と整合性のとれる透かし情報が混入された領域を指定する。
【００２９】
暗号抽出器５は、領域指定器４で指定された領域の画像データに対して、前記の暗号混入器３と同様の手順を行って暗号を抽出する。すなわち、まず、領域指定器４で指定された特定領域の画像信号の偶数ラインの輝度値の平均値を算出する。次に、この偶数ラインの画像信号の輝度値の平均値と特定領域の奇数ラインの画像信号の輝度値とを比較し、偶数ラインの画像信号の輝度値の平均値の方が大きい場合はカウンタＣｎｔ（ＥＶＥＮ）を１インクリメントし、偶数ラインの画像信号の輝度値の平均値の方が小さい場合はカウンタＣｎｔ（ＯＤＤ）を１インクリメントする。
【００３０】
こうして、特定領域の奇数ラインのすべての画素の輝度値と偶数ラインの画像信号の輝度値の平均値を比較し、その比較によりＣｎｔ（ＯＤＤ）＞Ｃｎｔ（ＥＶＥＮ）の場合は、暗号として”１”を得、逆にＣｎｔ（ＯＤＤ）＜Ｃｎｔ（ＥＶＥＮ）の場合は、暗号として”０”を得る。こうして得られた暗号情報は、暗号復号器６に入力され、復号化されて元の透かし情報を得る。この実施の形態では、特定領域が隣接する画像ラインであるので、水平方向のフィルタ等による帯域制限手段に対して、耐性の強い電子透かし情報の記録、再生ができる。
【００３１】
なお、本発明は上記の実施の形態に限定されるものではなく、例えば、上記の実施の形態では、電子透かし情報が埋め込まれた画像信号、すなわち、透かし埋め込み画像を記録媒体に記録し、再生するように説明したが、上記の透かし埋め込み画像をインターネットや有線回線あるいは無線回線などを介して配信し、配信された透かし埋め込み画像を受信して透かし情報を再生するようにしてもよい。
【００３２】
また、上記の実施の形態では、暗号混入器３及び暗号抽出器５において、奇数ラインの画像信号の輝度値を修正するように説明したが、偶数ラインの画像信号の輝度値を修正するようにしても構わない。
【００３３】
また、透かし情報を混入する画像信号の特定領域を連続する２ラインとしたが、領域指定器１及び領域指定器４において、整合性がとれていれば、任意の矩形領域や特殊な形状の領域を指定しても構わない。この場合、領域内の奇数ラインと偶数ラインをペアとして、上記と同様の暗号混入、抽出処理を行うことにより、１ビット以上の暗号を送ることが可能となるし、また、領域内の任意の１ラインの輝度の平均値と残りのラインの輝度値との比較とを行うことで暗号混入、抽出が可能になる。
【００３４】
更に、領域指定器１及び領域指定器４において、整合性がとれていれば、透かし情報を混入する画像信号の特定領域を画像内部に複数設けても構わない。更にまた、暗号発生器２における暗号化方式についても、暗号復号器６との整合性がとれていれば、どのような暗号化方式を用いても構わない。また、輝度値の修正の代わりに、特定領域内の任意領域の画素の色差値の平均値と、その任意領域以外の画素の色差値の大小を比較し、その比較結果に応じて電子透かし情報の埋め込みを行うようにしてもよい。
【００３５】
【発明の効果】
以上説明したように、本発明によれば、電子透かし情報を暗号化した暗号情報に応じて、入力画像信号中の特定領域内の第１の領域と第２の領域の各画素の輝度値又は色差値を、互いに相反する方向へ変換し、電子透かし情報が埋め込まれた画像信号として出力するようにしたため、元の入力画像信号の変化させる値を、第１の領域及び第２の領域のどちらか一方の各画素の輝度値又は色差値を変換する場合に比べて、約半分の値で済ませることも可能となり、入力画像信号からの画素当たりの変化量を少なくでき、よって、高画質な画像伝送を確保しつつ電子透かし情報を埋め込むことができる。
【００３６】
また、本発明によれば、特定領域として隣接する画像ラインを指定することにより、隣接する画像ライン間の相関性が非常に高いという性質を利用した電子透かし情報の埋め込みができるため、暗号情報の値に応じた特定領域の輝度値又は色差値の修正が比較的小さな値で済み、視覚特性上の画質劣化が認知しにくくでき、また、簡単に電子透かし情報を埋め込むことができる。
【００３７】
また、本発明によれば、特に特定領域を画像ラインとした場合、水平方向のフィルタ等による帯域制限手段に対して、耐性の強い電子透かし情報埋め込みや再生ができる。
【００３８】
更に、本発明によれば、電子透かし情報を埋め込んだ画像と入力画像との差分をとるなどの攻撃にあった場合にも、画像全体に差分がでてくるので、アルゴリズムを解析しにくくできる。
【図面の簡単な説明】
【図１】本発明装置の記録部の一実施の形態のブロック図である。
【図２】本発明装置により埋め込まれた電子透かし情報の再生部の一例のブロック図である。
【図３】本発明により暗号を混入する特定領域の一例を示す図である。
【図４】特定領域内の奇数ラインの画素値を電子透かし情報に応じて修正することを説明する図である。
【図５】本発明により、特定領域内の奇数ラインと偶数ラインの各画素値を、電子透かし情報に応じて相反する方向に修正することを説明する図である。
【符号の説明】
１、４　領域指定器
２　暗号発生器
３　暗号混入器
５　暗号抽出器
６　暗号復号器
１０　特定領域[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an apparatus and method for embedding digital watermark information, and more particularly to electronically embedding images, sounds, and multimedia data in order to prove copyright ownership, identify copyright infringement, and transmit hidden messages. The present invention relates to an electronic watermark information embedding device and an embedding method for embedding watermark information.
[0002]
[Prior art]
The digital watermark signal is a technique for embedding and hiding some information in multimedia data of images and sounds. There are various embedding methods. As a conventionally known method of embedding digital watermark information, for example, an information embedding method based on a code obtained by compressing and encoding information by the MPEG method, particularly a DCT coefficient, a motion vector, and a change in a quantization characteristic has been proposed. (Hiroshi Ogawa et al., "Method of Embedding Copyright Information in Digital Video Using DCT", SCIS '97 -31G).
[0003]
As another conventional digital watermark information embedding method, a method of spreading an image signal with a PN sequence according to a direct spreading method and synthesizing signature information with an image is also known (Junji Onishi et al., "PN Sequence" Watermark Signature Method for Images ", SCIS '97_26B). In this conventional method, when an image signal containing a signature is despread, the signature information is spread over the entire image signal, and the spread signal is very weak, does not cause a large noise to the image signal, and the signature information is The included image signal is apparently the same as the original image. To confirm the signature information, the signal of the signature information is detected by spreading the image signal with a spreading code.
[0004]
[Problems to be solved by the invention]
However, in the above-mentioned conventional digital watermark information embedding method, when watermark information is embedded in an image and sent, there is a problem that it takes time and cost to convert the image. Further, when watermark information is embedded in a transform coefficient calculated by using an orthogonal transform such as DCT, there is a problem that the watermark information may be lost even with a simple low-pass filter.
[0005]
The present invention has been made in view of the above points, and has as its object to provide an electronic watermark information embedding apparatus and an embedding method that can easily embed electronic watermark information in an image.
[0006]
It is another object of the present invention to provide a digital watermark information embedding device and a digital watermark information embedding method capable of recording and reproducing digital watermark information having high resistance to a filter.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, a digital watermark information embedding device according to the present invention includes an area designating unit for designating a specific area in which digital watermark information in an input image signal is embedded, and encrypting the digital watermark information as encryption information. An encryption generating means for outputting, and an average value of luminance values or color difference values of all pixels in the first area in the specific area in the input image signal and a second area other than the first area in the specific area. By comparing the brightness value or the color difference value of each pixel with each other, a first counter value indicating the number of pixels whose brightness value or color difference value is larger than the average value among all the pixels in the second region, and a brightness value or A second counter value indicating the number of pixels whose color difference value is smaller than the average value is obtained, and the magnitude relation between the first and second counter values is determined by the first value or the second value of the encryption information from the encryption generating means. The relationship is set in advance according to the value. And the region and converting the luminance value or color difference values of each pixel in the second region in the opposite direction, in which a structure having a mixing means for outputting an image signal in which the electronic watermark information has been embedded.
[0008]
According to the present invention, the luminance value or the color difference value of each pixel of the first region and the second region of the specific region in the input image signal is converted in the opposite direction according to the encryption information obtained by encrypting the digital watermark information. Since the digital watermark information is output as an embedded image signal, by specifying adjacent image lines as the first and second areas, the property that the correlation between the adjacent image lines is extremely high is obtained. The used digital watermark information can be embedded.
[0009]
Here, the mixing means includes an average value calculating means for calculating an average value of luminance values or color difference values of pixels in a first area in the specific area in the input image signal; A first counter value indicating the number of pixels whose luminance value or chrominance value is larger than the average value among all the pixels in the second region. Counter value calculating means for calculating a second counter value indicating the number of pixels whose luminance value or color difference value is smaller than the average value; counter value comparing means for comparing the first and second counter values; When the value of the encryption information from the encryption generating means is the first value, a comparison result in which the first counter value is larger than the second counter value is obtained by the counter value comparison means, and the encryption information from the encryption generation means is obtained. When the value of the second value is the second value, Conversion means for converting the luminance values or the color difference values of all the pixels in the first area and the second area in opposite directions so that a comparison result in which the first counter value is smaller than the second counter value is obtained. And outputs a signal obtained by converting the luminance value or color difference value of each pixel in the first area and the second area according to the value of the encryption information by the conversion means as an image signal in which digital watermark information is embedded. It is characterized by doing.
[0010]
In order to achieve the above object, an embedding method according to the present invention includes a first step of designating a specific area in which digital watermark information in an input image signal is to be embedded, and encrypting the digital watermark information as encryption information. A second step of outputting, and an average value of luminance values or color difference values of all pixels of the first area in the specific area in the input image signal and a second area other than the first area in the specific area A first counter value indicating the number of pixels whose luminance value or color difference value is larger than the average value among all pixels in the second area, Alternatively, a third step of obtaining a second counter value indicating the number of pixels whose color difference value is smaller than the average value, and the magnitude relationship between the first and second counter values may be determined by determining the first value or the second value of the encryption information. The first region is set so as to have a relationship set in advance according to the value. The luminance values or chrominance values of each pixel of the second area is converted in opposite directions, characterized in that it comprises a fourth step of outputting an image signal in which the electronic watermark information has been embedded.
[0011]
According to the present invention, the luminance value or the color difference value of each pixel of the first region and the second region of the specific region in the input image signal is converted in the opposite direction according to the encryption information obtained by encrypting the digital watermark information. Since the digital watermark information is output as an embedded image signal, by specifying adjacent image lines as the first and second areas, the property that the correlation between the adjacent image lines is extremely high is obtained. The used digital watermark information can be embedded.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a recording unit of a digital watermark recording / reproducing apparatus according to the present invention. In the figure, an image signal is supplied to an area designator 1 where an area in which digital watermark information is embedded is determined.
[0013]
On the other hand, the encryption generator 2 encrypts digital watermark information embedded in an image signal. As a simple example during encryption, when an ASCII character is sent, the character is represented by one byte (8 bits), so that eight bits may be sent one by one in order. As long as compatibility with the encryption / decryption unit 6 of the digital watermark information reproducing unit in FIG. 2 can be obtained, data encoded using a table such as a Huffman table may be transmitted bit by bit.
[0014]
The encryption mixing unit 3 converts the luminance value of the image signal into a specific area of the image signal specified by the area specification unit 1 based on the encryption information sent from the encryption generator 2. Here, an area 10 of the upper two lines of the image signal, which is indicated by a dotted line in FIG. 3, is defined as a specific area (in FIG. 3, white circles indicate pixels of odd-numbered lines, and black circles indicate pixels of even-numbered lines). The number of effective pixels per line is N pixels.
[0015]
First, the average value P _AVE (EVEN) of the luminance values of the image signals of the even-numbered lines in the odd-numbered lines and the even-numbered lines of the specific area 10 is calculated by the following equation.
[0016]
(Equation 1)

Here, in the expression (1), Pi (EVEN) indicates the luminance value of the image signal of the even-numbered line in the specific area 10.
[0017]
Next, the calculated average value P _AVE (EVEN) of the luminance values of the image signals of the even-numbered lines is compared with the luminance value Pi (ODD) of the image signals of the odd-numbered lines in the specific region 10, and the inequality Pi ( ODD) -P _AVE (EVEN)> 0 (i = 1, 2,..., N) (2)
There When satisfied, the positive counter CNTposi incremented by one, conversely _{Pi (ODD) -P AVE (EVEN} ) <0 (i = 1,2, .., N) (3)
Holds, the negative counter CNTnega is incremented by one. This comparison is performed for the luminance values of all the pixels on the odd-numbered line in the specific area.
[0018]
Here, the meaning of the encryption information in the digital watermark information recording unit and the digital watermark information reproducing unit is set in advance. In this case, when the bit of the encryption information to be embedded is “1”,
CNTposi> CNTnega (4)
And when the bit of the encryption information to be embedded is “0”,
CNTposi <CNTnega (5)
Then, the luminance value of the image signal of the odd line and the luminance value of the image signal of the even line are corrected in a direction opposite to the direction in which the luminance value of the image signal of the odd line is corrected.
[0019]
In order to satisfy the expression (4) (to embed the bit “1” of the encryption information), the luminance level of the image signal of the odd-numbered line is all increased by one, and the image of the even-numbered line is again increased. A comparison is made with the average value P _AVE (EVEN) of the luminance values of the signals, and if the expression (4) is satisfied, the processing is terminated. However, if the equation (4) is still not satisfied, the luminance levels of the image signals of the odd-numbered lines are all increased by one again, and the average value P _AVE (EVEN) of the luminance values of the image signals of the even-numbered lines is again increased. ) Is repeated until Expression (4) is satisfied.
[0020]
At this time, in a pixel whose luminance value level increases by 2 or more, the even number of the increment operation is opposite to the direction of correcting the luminance value of the image signal on the even line side and the luminance value of the image signal on the odd line side. By performing the correction in the direction in which the luminance value of the image signal of the even-numbered line is relatively increased, the luminance value of the image signal of the odd-numbered line is relatively equal to twice when viewed from the luminance value of the image signal of the even-numbered line.
[0021]
Similarly, when the bit “0” of the encryption information is embedded, the luminance value of the image signal of the odd-numbered line is corrected by lowering the luminance level by one, and then the luminance value of the image signal of the even-numbered line is again adjusted. The comparison with the average value P _AVE (EVEN) is repeated until Expression (5) is satisfied.
[0022]
At this time, in a pixel whose luminance value level decreases by 2 or more, the even-numbered operation of the decrementing operation is opposite to the direction in which the luminance value of the image signal on the even-numbered line side is corrected and the luminance value of the image signal on the odd-numbered line side is corrected. By performing the correction in such a direction, the luminance value of the image signal of the odd-numbered line is relatively decremented twice when viewed from the luminance value of the image signal of the even-numbered line.
[0023]
In the above method, since the correlation between adjacent image lines is extremely high, the correction of the luminance level can be made relatively small, so that the image quality deterioration in the visual characteristics can be suppressed to a level that is difficult to recognize, and the encryption can be easily performed. Embedding of information can be realized. In this manner, an image signal in which the luminance value of each pixel of the odd-numbered line in the specific area is corrected based on the bit of the encryption information is output from the encryption mixing unit 3 as an electronic watermark embedded image signal, and is output by a known recording unit (not shown). It is recorded on a recording medium.
[0024]
Here, a case where only the pixel value of one specific region is corrected is shown in FIG. That is, FIG. 4 shows a case where it is assumed that the luminance value of each pixel of only the odd-numbered line is corrected. Assuming that all the data of the even-numbered line is assumed to be a reference value by a dotted line, only the data of the odd-numbered line is used. For example, an example of +2 or -2 is shown. The pixel correction value in FIG. 4 conceptually indicates that it may be a luminance value of a specific one pixel or an average luminance value of a line.
[0025]
On the other hand, in the present invention, instead of correcting only the luminance value of one line of the specific area, as shown in FIG. 5, the luminance values of both the odd and even lines of the specific area are corrected in opposite directions. I do. That is, the luminance value difference between the even-numbered line and the odd-numbered line is relatively the same as that in FIG. 4, but by correcting both lines, only one of the odd-numbered and even-numbered lines (the odd-numbered line in FIG. 4) is obtained. Can be corrected to half the value to be changed.
[0026]
The method of correcting the luminance values of the odd-numbered lines and the even-numbered lines in opposite directions is used when at least two or more changes in the luminance values are assumed when only one of the lines is used. That is, as compared with the case where the luminance value of each pixel of only the odd line is corrected as shown in FIG. 4, the case where the luminance value of each pixel of the odd line and the even line is corrected in the opposite direction is shown in FIG. Become like
[0027]
In FIG. 5, the relative difference between the data of the even-numbered line and the data of the odd-numbered line is made the same, and the center of the difference is assumed as a reference value by a dotted line, and the data of the odd-numbered line is, for example, +1 or -1, This shows an example in which the data of the even line is incremented in the opposite direction, for example, by −1 or +1. The pixel correction value in FIG. 5 conceptually indicates that it may be a luminance value of one specific pixel or an average luminance value of a line.
[0028]
Next, the configuration and operation of the playback unit will be described. FIG. 2 is a block diagram showing an embodiment of the reproducing unit of the digital watermark recording / reproducing apparatus. In the figure, a digital watermark embedded image signal reproduced from a recording medium by a known reproducing unit is input to an area designator 4, where watermark information compatible with the area designator 1 of the digital watermark information recording unit is obtained. Specify the mixed area.
[0029]
The cipher extractor 5 extracts the cipher from the image data in the area specified by the area specifier 4 by performing the same procedure as that of the cipher mixer 3. That is, first, the average value of the luminance values of the even lines of the image signal in the specific area designated by the area designator 4 is calculated. Next, the average value of the luminance values of the image signals of the even lines is compared with the luminance value of the image signals of the odd lines in the specific area. If the average value of the luminance values of the image signals of the even lines is larger, the counter is determined. Cnt (EVEN) is incremented by one, and when the average value of the luminance values of the image signals of the even lines is smaller, the counter Cnt (ODD) is incremented by one.
[0030]
In this way, the luminance values of all the pixels on the odd-numbered lines in the specific area are compared with the average value of the luminance values of the image signals on the even-numbered lines. If Cnt (ODD)> Cnt (EVEN) as a result of the comparison, "1" is ciphered. , And conversely, if Cnt (ODD) <Cnt (EVEN), “0” is obtained as encryption. The encryption information thus obtained is input to the encryption / decryption unit 6, and is decrypted to obtain the original watermark information. In this embodiment, since the specific area is an adjacent image line, it is possible to record and reproduce digital watermark information that is resistant to band limiting means such as a filter in the horizontal direction.
[0031]
The present invention is not limited to the above embodiment. For example, in the above embodiment, an image signal in which digital watermark information is embedded, that is, a watermark embedded image is recorded on a recording medium and reproduced. However, the watermark embedded image may be distributed via the Internet, a wired line, a wireless line, or the like, and the distributed watermark embedded image may be received to reproduce the watermark information.
[0032]
Further, in the above embodiment, the description has been made such that the luminance value of the image signal of the odd line is corrected in the encryption mixing unit 3 and the encryption extractor 5, but the luminance value of the image signal of the even line is corrected. It does not matter.
[0033]
Although the specific area of the image signal into which the watermark information is mixed is two continuous lines, if the

area designators

1 and 4 are consistent, any rectangular area or area having a special shape can be used. May be specified. In this case, the odd-numbered lines and the even-numbered lines in the area are paired, and the same encryption and extraction processing as described above is performed, so that it is possible to send a 1-bit or more encryption. By comparing the average value of the luminance of one line with the luminance value of the remaining lines, it is possible to mix and extract the code.
[0034]
Furthermore, as long as consistency is ensured in the

area designators

1 and 4, a plurality of specific areas of the image signal into which the watermark information is mixed may be provided in the image. Furthermore, as for the encryption method in the encryption generator 2, any encryption method may be used as long as the encryption method is compatible with the encryption / decryption device 6. Also, instead of correcting the luminance value, the average value of the color difference values of the pixels in the arbitrary region in the specific region is compared with the color difference values of the pixels other than the arbitrary region, and the digital watermark information is determined according to the comparison result. May be embedded.
[0035]
【The invention's effect】
As described above, according to the present invention, the luminance value or the luminance value of each pixel of the first area and the second area in the specific area in the input image signal according to the encryption information obtained by encrypting the digital watermark information. Since the color difference values are converted in directions opposite to each other and output as an image signal in which digital watermark information is embedded, the value to be changed of the original input image signal can be changed in either the first area or the second area. As compared with the case of converting the luminance value or the color difference value of each one of the pixels, it is possible to use only about half the value, and the amount of change per pixel from the input image signal can be reduced. Digital watermark information can be embedded while ensuring transmission.
[0036]
Further, according to the present invention, by designating adjacent image lines as a specific area, it is possible to embed digital watermark information utilizing the property that the correlation between adjacent image lines is extremely high. The correction of the luminance value or the color difference value of the specific area according to the value may be a relatively small value, and it is possible to make it difficult to perceive image quality deterioration in visual characteristics, and to easily embed digital watermark information.
[0037]
Further, according to the present invention, especially when a specific area is an image line, it is possible to embed or reproduce digital watermark information that is highly resistant to band limiting means such as a horizontal filter.
[0038]
Furthermore, according to the present invention, even in the case of an attack such as obtaining a difference between an image in which digital watermark information is embedded and an input image, a difference appears in the entire image, so that it is difficult to analyze the algorithm.
[Brief description of the drawings]
FIG. 1 is a block diagram of an embodiment of a recording unit of the apparatus of the present invention.
FIG. 2 is a block diagram of an example of a digital watermark information reproducing unit embedded by the apparatus of the present invention.
FIG. 3 is a diagram showing an example of a specific area into which encryption is mixed according to the present invention.
FIG. 4 is a diagram illustrating that pixel values of an odd line in a specific area are corrected according to digital watermark information.
FIG. 5 is a diagram for explaining that each pixel value of an odd-numbered line and an even-numbered line in a specific area is corrected in opposite directions according to digital watermark information according to the present invention.
[Explanation of symbols]
1, 4 area designator 2 encryption generator 3 encryption mixing unit 5 encryption extraction unit 6 encryption / decryption unit 10 specific area

Claims

Area designating means for designating a specific area in which digital watermark information in the input image signal is to be embedded,
Encryption generating means for encrypting the digital watermark information and outputting it as encryption information;
The average value of the luminance values or the color difference values of all the pixels in the first area in the specific area in the input image signal and the average value of the pixels in the second area other than the first area in the specific area By comparing the brightness value or the color difference value with each other, a first counter value indicating the number of pixels whose brightness value or color difference value is larger than the average value among all the pixels in the second area, and the brightness value or A second counter value indicating the number of pixels whose color difference value is smaller than the average value is obtained, and the magnitude relationship between the first and second counter values is determined by the first value or the first value of the encryption information from the encryption generation unit. The luminance value or the color difference value of each pixel in the first area and the second area is converted in the opposite direction so that the relationship is set in advance according to the second value, and the digital watermark information is embedded. And a mixing means for outputting as an image signal. And information embedding apparatus.

The mixing unit includes: an average value calculation unit configured to calculate an average value of a luminance value or a color difference value of a pixel in a first region in the specific region in the input image signal; The first and second pixels are compared with each other in the second area other than the first area, and among all the pixels in the second area, the first number indicating the number of pixels whose luminance value or color difference value is larger than the average value Counter value calculating means for calculating a counter value of the second counter value and a second counter value indicating the number of pixels whose luminance value or color difference value is smaller than the average value, and comparing the first and second counter values with each other. A counter value comparison unit, and a comparison result in which the first counter value is larger than the second counter value by the counter value comparison unit when the value of the encryption information from the encryption generation unit is a first value. Is obtained from the cipher generation means. When the value of the cryptographic information is the second value, the first area and the first area are compared so that the counter value comparing means can obtain a comparison result in which the first counter value is smaller than the second counter value. Conversion means for converting the luminance values or the color difference values of all the pixels in the second area in opposite directions; and the conversion means converts the luminance value or the color difference of each pixel in the first area and the second area. 2. The digital watermark information embedding device according to claim 1, wherein a signal whose value is converted according to the value of the encryption information is output as an image signal in which the digital watermark information is embedded.

A first step of designating a specific area in which digital watermark information in the input image signal is to be embedded;
A second step of encrypting the digital watermark information and outputting it as encryption information; and an average value of luminance values or color difference values of all pixels of a first area in the specific area in the input image signal. A luminance value or a color difference value of each pixel in a second region other than the first region in the specific region is compared in magnitude with each other, and among all the pixels in the second region, the luminance value or the color difference value is the average. A third step of obtaining a first counter value indicating the number of pixels larger than the value, and a second counter value indicating the number of pixels whose luminance value or color difference value is smaller than the average value;
The first area and the second area are set so that the magnitude relation between the first and second counter values is set in advance according to the first value or the second value of the encryption information. Converting the luminance value or color difference value of each pixel in the opposite direction and outputting the image signal as an image signal in which digital watermark information is embedded.

The third step is a fifth step of calculating an average value of luminance values or color difference values of pixels of a first area in the specific area in the input image signal; Compare the size of each pixel in the second area other than the first area in the area, and indicate the number of pixels having a luminance value or a color difference value larger than the average value among all the pixels in the second area. A sixth step of calculating a first counter value and a second counter value indicating the number of pixels in which the luminance value or the color difference value is smaller than the average value, and the fourth step includes: A seventh step of comparing the first and second counter values with each other, and when the value of the cryptographic information is the first value, a comparison result in which the first counter value is larger than the second counter value is obtained. And when the value of the cryptographic information is a second value, the first An eighth step of converting the luminance values or the color difference values of all the pixels in the first area and the second area in opposite directions so as to obtain a comparison result in which the counter value is smaller than the second counter value. 4. The digital watermark information embedding method according to claim 3, further comprising the steps of: