JP4443009B2 - Information distribution system and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information distribution system capable of obtaining the original image and sound from data that are once delivered or distributed as a sample, etc. SOLUTION: In a server computer 100, an electronic watermark embedding part 104a embeds an electronic watermark in image data in a visible state. An image data transmitting part 104b transmits the image data. In a client computer 200, an image data receiving part 204a receives the image data. When the server computer 100 recognizes image purchase due to charging completion information from a charging management computer 350, a state alteration information transmitting part 104c transmits state alteration information. In the client computer 200, a state alteration information receiving part 204b receives the state alternation information, and a state alteration part 204c eliminates the electronic watermark embedded in the image data or alters the electronic watermark to an invisible state on the basis of the state alteration information.

Description

【００７５】
一方、画素ｗijの値が１（即ち、マーク部分）である場合には、その画素ｗijに対応するカラー画像Ｐの画素ｐijを選択し、その画素ｐijのＲＧＢ成分を式（２）に示す変換行列Ｂによって変換し、得られたＲＧＢ成分を署名済み画像Ｑの対応する画素ｑijのＲＧＢ成分とする（ステップＳ１２５）。
【００７６】
【数２】

【００７７】
即ち、画素ｗijの値が１の場合は、式（３）のように表される。
【００７８】
【数３】

【００７９】
式（２），（３）に示すように、変換行列Ｂは、３次の正方行列であって、対角成分の全てが値ｚ（但し、ｚは任意の実数）から成り、その他の成分が値ｘ（但し、ｘは０以外の任意の実数）または−ｘから成る。そして、各行及び各列の成分の和はそれぞれｚとなっている。このうち、対角成分である値ｚは、各画素の輝度成分に対して著しい影響を与えるため、本明細書では、この値ｚを明るさ係数と呼ぶ。また、残りの係数ｘは、透かし情報を生成するための秘密の係数であり、本明細書では、この値ｘを透かし係数と呼ぶ。この透かし係数の演算結果によって透かし信号となる色差信号が得られる。
【００８０】
但し、前述したように、本実施例において、透かし画像Ｗ、即ち、電子透かしは、可視状態で埋め込むようにするため、変換行列Ｂにおける明るさ係数ｚを１以外の値に設定する。すると、式（３）に基づいて、署名済み画像Ｑにおけるマーク部分の画素のＲＧＢ成分は、式（４）の如くになる。
【００８１】
【数４】

【００８２】
式（４）の右辺を見ると、第１項では、カラー画像Ｐにおけるマーク部分の画素のＲＧＢ成分に、それぞれ明るさ係数ｚが掛けられており、この明るさ係数ｚによって３原色の輝度成分に均等に影響を与えることになる。また、第２項では、透かし信号である色差信号が新たに加算されることになる。この結果、明るさ係数ｚに基づいて透かし情報が可視状態で埋め込まれると共に、透かし係数ｘに基づいた透かし情報が潜在的に埋め込まれることになる。
【００８３】
ここで、明るさ係数ｚを正の実数値と設定すると、０＜ｚ＜１の範囲では、マーク部分を黒色系に表示することができ、ｚ＞１の範囲では、マーク部分を白色系に表示することが可能である。また、この時、明るさ係数ｚの値が１に近いほど、マーク部分の輝度が薄くなり、不可視状態に近づく。一方、明るさ係数ｚの値が１から離れるに従い、マーク部分には一層はっきりとした輝度が得られる。
【００８４】
なお、透かし画像を可視状態で埋め込む際には、輝度を大きく変化させることにより、画素値が上限値を超えてオーバフローとなったり、下限値を超えてアンダフローとなったりしやすいので、注意しなければならない。もし、そのようなオーバフローやアンダフローが生じた場合には、その画素値を強制的に上限値または下限値に置き換えることにより、補正するようにする。
【００８５】
また、変換行列Ｂによって変換した結果、得られたＲＧＢ成分が整数でない場合には、小数点以下を四捨五入して整数化する。
【００８６】
以上のような演算処理を行なった後、変数ｉ，ｊを順次インクリメントしつつ（ステップＳ１３５，Ｓ１４５）、透かし画像Ｗの全画素について処理が完了したか否かを判断し（ステップＳ１４０，Ｓ１５０）、完了するまで、カラー画像Ｐに電子透かしである透かし画像Ｗを埋め込む上記処理を繰り返す。全画素についての処理が完了すれば、透かし画像の埋め込み処理を終了する。
【００８７】
ここで、カラー画像Ｐに電子透かしである透かし画像Ｗを可視状態で埋め込んだ場合の具体例について説明する。
【００８８】
この具体例では、対象となるカラー画像Ｐとして、図５（ａ）に示すポートレート画像と、図５（ｂ）に示すカフェテリア画像を用いた。これらの画像は日本規格協会から発行されているカラーデジタル標準画像データを縮小したもので、大きさが３２０×２５６画素、ＲＧＢ成分が各々８ビットの画像である。一方、埋め込むべき電子透かしである透かし画像Ｗとしては、図６に示すロゴマーク画像を用いた。
【００８９】
図６に示す可視用のロゴマーク画像を、図５（ａ）に示すポートレート画像に埋め込む際には、黒色系表示を採用するために、明るさ係数ｚを０．７に設定すると共に、透かし係数ｘを０．１に設定するようにした。しかし、図５（ｂ）に示すカフェテリア画像に埋め込む際には、ポートレート画像の場合とロゴマークの明るさを変えて、よりはっきりとロゴマークを認識できるようにするために、明るさ係数ｚを０．４に設定した。なお、透かし係数ｘは同じ０．１に設定した。
【００９０】
このような埋め込みの結果、各署名済み画像は図７（ａ），（ｂ）に示すように表される。図７（ａ）と（ｂ）を比較すれば明らかなように、可視用のロゴマークは、明るさ係数ｚの値によって、異なる輝度で表示させることが可能である。
【００９１】
以上のようにして、署名済み画像、即ち、電子透かしが可視状態で埋め込まれた画像が得られたら、次に、サーバコンピュータ１００のＣＰＵ１０２は、その画像の存在を、Ｗｅｂページや電子メールなどの手段を利用して、ネットワーク３００を介してクライアントコンピュータ２００に知らせる（ステップＳ１４）。その知らせを受けて、クライアントコンピュータ２００のユーザが、その画像の取得を希望すれば、そのユーザの指示によって、クライアントコンピュータ２００のＣＰＵ２０２は、その画像のダウンロードをサーバコンピュータ１００に対して要求する（ステップＳ１６）。
【００９２】
サーバコンピュータ１００では、その要求を受け取ると、画像データ送信部１０４ｂが、電子透かしが可視状態で埋め込まれた画像データをハードディスク１１０から読み出して、通信装置１２０からネットワーク３００を介してクライアントコンピュータ２００に送信する（ステップＳ１８）。クライアントコンピュータ２００では、画像データ受信部２０４ａが、サーバコンピュータ１００から送信された画像のデータを通信装置２２０を介して受信し、ハードディスク２１０に格納する（ステップＳ２０）。
【００９３】
その後、クライアントコンピュータ２００のユーザが、その画像の表示を指示すると、クライアントコンピュータ２００のＣＰＵ２０２は、その画像データをハードディスク２１０から読み出して表示装置２１２に表示させる。ユーザは、これにより、画像の内容は十分把握できるものの、このとき表示される画像には、電子透かしが可視状態で埋め込まれているため、図７（ａ），（ｂ）に示したとおり、電子透かしが画像の邪魔になり、このまま、この画像を利用するのには不都合となっている。すなわち、この状態では、この画像は、ユーザに対してサンプル用として提供されていることになる。
【００９４】
従って、ユーザが、この画像を気に入って、この画像を不都合なく利用することを希望する場合には、可視状態で埋め込まれている電子透かしを見えなくするために、この画像を購入する必要がある。
【００９５】
そこで、ユーザが、この画像の購入を決定し、その旨をクライアントコンピュータ２００に指示すると、クライアントコンピュータ２００は、画像購入の指示をネットワーク３００を介して課金管理コンピュータ３５０に送る（ステップＳ２２）。課金管理コンピュータ３５０では、その指示を受け取ると、内部にある課金管理情報にアクセスして、クライアントコンピュータ２００のユーザを特定し、そのユーザに対して、画像購入に対する課金を行う（ステップＳ２４）。
【００９６】
なお、このとき、そのユーザが新規ユーザである場合には、課金管理コンピュータ３５０は、課金処理に先立って、クライアントコンピュータ６００に対しユーザ登録手続を行うべき旨を指示する。そして、課金管理コンピュータ３５０は、クライアントコンピュータ６００のユーザによりユーザ登録手続が行われたことを確認したならば、その後、上記した課金処理を行う。
【００９７】
そして、課金管理コンピュータ３５０は、そのユーザに対する課金完了の情報をネットワーク３００を介してサーバコンピュータ１００に送る（ステップＳ２６）。サーバコンピュータ１００では、その情報を受け取ると、クライアントコンピュータ２００のユーザが、配信した画像を購入したことを認識する。
【００９８】
ところで、可視状態で埋め込んだ電子透かしを見えなくする場合、電子透かしを画像データから除去してしまう方法と、電子透かしを画像データ内に留めたまま、視覚では認識し難い状態、即ち、不可視状態に変更する方法と、が考えられる。前者の方法による場合には、画像データから電子透かしが完全になくなるので、その画像が不正利用された場合に、その発見や追跡が困難になってしまう。従って、後者の方法を採った方が、不正利用防止の観点からも望ましい。
【００９９】
よって、本実施例では、可視状態で埋め込まれた電子透かしを不可視状態に変更する後者の方法について、まず説明する。
【０１００】
そこで、まず、サーバコンピュータ１００では、画像購入を認識したら、状態変更情報送信部１０４ｃが、可視状態にある電子透かしを不可視状態にするための状態変更情報を、ネットワーク３００を介してクライアントコンピュータ２００に送信する（ステップＳ２８）。
【０１０１】
状態変更情報とは、具体的には、前述のステップＳ１２５で用いた変換行列Ｂにおける明るさ係数ｚと透かし係数ｘの各値である。これらの値は、電子透かしを埋め込んだ者のみが知る秘密鍵であり、この値を用いることによって、可視状態にある電子透かしを不可視状態に変更することが可能となる。
【０１０２】
そこで、状態変更情報送信部１０４ｃは、この秘密鍵である状態変更情報を第三者に知られないようにするために、暗号化した上で、クライアントコンピュータ２００に送信する。
【０１０３】
なお、状態変更情報の伝送には、サーバコンピュータ１００からの電子メールを利用してもよいし、クライアントコンピュータ２００からのダウンロードによって行なうようにしてもよい。
【０１０４】
クライアントコンピュータ２００では、状態変更情報受信部２０４ｂが、送信されきた状態変更情報を受信し、暗号化されているその状態変更情報を復号化する（ステップＳ３０）。そして、状態変更部２０４ｃが、ハードディスク２１０から電子透かしの埋め込まれている画像データを読み出し、復号化した状態変更情報に基づいて、その電子透かしを可視状態から不可視状態に変更する処理を行う（ステップＳ３２）。
【０１０５】
具体的には、状態変更部２０４ｃは、図８に示す状態変更処理の手順に従って、電子透かしの状態変更を行う。図８は図１の状態変更部２０４ｃが行う状態変更処理の手順を示すフローチャートである。状態変更部２０４ｃは、まず、受信した状態変更情報から、変換行列Ｂにおける明るさ係数ｚと透かし係数ｘの各値を特定する処理を行なう（ステップＳ３０５）。
【０１０６】
次に、特定した明るさ係数ｚと透かし係数ｘの各値を用いて、変換行列Ｂの逆行列Ｂ^-1を生成する（ステップＳ３１０）。
【０１０７】
そして、電子透かしの埋め込まれた画像、すなわち、署名済み画像Ｑを特定する処理を行なう（ステップＳ３１５）。この処理は、ハードディスク２１０に記憶された画像ファイルを指定することにより行なわれる。
【０１０８】
続いて、画像処理用の変数ｉ，ｊを初期値０にセットする（ステップＳ３２０，３２５）。そして、その変数ｉ，ｊについて、署名済み画像Ｑの画素ｑijが可視のマーク部分の画素であるか否かの判断を行なう（ステップＳ３３０）。
【０１０９】
判断の結果、画素ｑijが、非マーク部分の画素である場合には、その画素ｑijのＲＧＢ成分を、そのまま、新たな署名済み画像Ｑ’の対応する画素ｑij’のＲＧＢ成分とする（ステップＳ３４０）。
【０１１０】
一方、画素ｑijが可視のマーク部分の画素である場合には、その画素ｑijのＲＧＢ成分を、ステップＳ３１０で生成した逆行列Ｂ^-1によって、式（５）に従って変換して、原画像であるカラー画像Ｐの対応する画素ｐijのＲＧＢ成分を得る（ステップＳ３３５）。
【０１１１】
【数５】

【０１１２】
但し、実際には、図３に示した電子透かしの埋め込み処理において、カラー画像Ｐにおける画素ＰijのＲＧＢ成分を変換行列Ｂで変換した際に、その変換結果が整数でない場合には、小数点以下を四捨五入して整数化しているため、上記したように、ステップＳ３３５で、署名済み画像Ｑにおける画素ｑijのＲＧＢ成分を、変換行列のＢの逆行列Ｂ^-1で変換したとしても、得られる値は、原画像であるカラー画像Ｐにおける画素ｐijの元のＲＧＢ成分と必ずしも一致しない。即ち、ステップＳ３３５の処理では、原画像であるカラー画像Ｐにおける画素ｐijの元のＲＧＢ成分の近似値を得ることになる。
【０１１３】
次に、得られた画素ｐijのＲＧＢ成分を、明るさ係数ｚを１に、透かし係数ｘはステップＳ３０５で特定した値のまま（即ち、先の埋め込み処理時と同じ値）に、それぞれ設定した変換行列Ｂによって、前述した式（３）に従って変換し、得られたＲＧＢ成分を新たな署名済み画像Ｑ’の対応する画素ｑij’のＲＧＢ成分とする（ステップＳ３４５）。
【０１１４】
即ち、この場合は、式（６）のように表される。
【０１１５】
【数６】

【０１１６】
ここで、Ｂ₁は、明るさ係数ｚを１に設定した変換行列Ｂを表し、ｑij¹は、変換行列Ｂ₁で変換して得られた画素のＲＧＢ成分を表す。具体的に、ｑij¹のＲＧＢ成分は、式（７）に示す如く表される。
【０１１７】
【数７】

【０１１８】
式（７）の右辺を見ると、第１項では、原画像であるカラー画像Ｐにおける画素ｐijのＲＧＢ成分（実際にはの近似値）がそのまま維持されることになり、第２項では、透かし信号である色差信号のみが新たに加算されることになる。この結果、各画素の輝度成分はそのまま維持されることになり、透かし信号として透かし係数ｘに基づく透かし情報のみが埋め込まれることになり、よって、透かし情報が不可視状態で埋め込まれることになる。
【０１１９】
つまり、ステップＳ３３５及びＳ３４５の処理を経ることによって、可視のマーク部分であった画素は、不可視状態に変更されて、不可視のマーク部分の画素になる。
【０１２０】
なお、ステップ３４５においても、ステップＳ１２５の場合と同様に、変換行列Ｂによって変換した結果、得られたＲＧＢ成分が整数でない場合には、小数点以下を四捨五入して整数化する。
【０１２１】
以上のような演算処理を行なった後、変数ｉ，ｊを順次インクリメントしつつ（ステップＳ３５０，Ｓ３６０）、透かし画像Ｗの全画素について処理が完了したか否かを判断し（ステップＳ３５５，Ｓ３６５）、完了するまで、可視状態にある電子透かしを不可視状態に変更させる上記処理を繰り返す。全画素についての処理が完了すれば、電子透かしの状態変更処理を終了する。
【０１２２】
その後、クライアントコンピュータ２００のユーザが、画像の表示を指示すると、クライアントコンピュータ２００のＣＰＵ２０２は、電子透かしが不可視状態となった画像データをハードディスク２１０から読み出し、表示装置２１２に表示させる。このとき表示される画像には、電子透かしが一見すると除去されており、従って、ユーザは、この画像を不都合なく利用することが可能となる。
【０１２３】
ここで、可視状態で埋め込まれた電子透かしを不可視状態にする場合の具体例について簡単に説明する。
【０１２４】
図７（ａ），（ｂ）に示した署名済み画像について、可視のマーク部分を、図８に示した手順で不可視状態にした場合、結果としては、図９（ａ），（ｂ）に示すような新たな署名済み画像が得られる。図９を見れば明らかなように、可視状態で表示されていたマーク部分は不可視状態となって、一見すると、マーク部分が除去されたかのように見える。そのため、クライアントコンピュータ２００のユーザは、購入した画像を不都合なく利用することができる。
【０１２５】
但し、電子透かしは不可視状態になっただけで、画像から完全に除去されたわけではないので、仮に、この画像が不正利用された場合でも、その発見や追跡を容易に行うことができる。
【０１２６】
即ち、この画像が不正利用されたという疑いがある時には、図５（ａ），（ｂ）に示す原画像であるカラー画像を用いて、図９（ａ），（ｂ）に示す署名済み画像から、不可視状態にあるマーク部分を、所定の手順で抽出する。具体的には、原画像であるカラー画像と署名済み画像と差分を、ＲＧＢ各成分についてとり、その差分を２値化することによって、マーク部分を抽出する。抽出の結果としては、図１０（ａ），（ｂ）に示すようなロゴマーク画像が得られる。図１０を見れば明らかなように、不可視状態で埋め込まれていたロゴマーク画像を、痕跡が十分確認できる程度に抽出することができる。
【０１２７】
ところで、前述したように、可視状態で埋め込んだ電子透かしを見えなくする方法としては、電子透かしを画像データから除去してしまう方法と、電子透かしを不可視状態に変更する方法と、があり、上記の説明では、後者の方法を用いる場合について説明したが、前者の方法を用いる場合には、後者の方法における処理を次のように変更する。
【０１２８】
即ち、前述した後者の方法では、図８に示した状態変更処理において、原画像であるカラー画像Ｐの対応する画素ｐijのＲＧＢ成分を得た（ステップＳ３３５）後、電子透かしを不可視状態に変更するために、画素ｐijのＲＧＢ成分を、明るさ係数ｚを１に、透かし係数ｘを特定した値に、それぞれ設定した変換行列Ｂによって変換する（ステップＳ３４５）ようにしていたが、前者の方法では、このステップＳ３４５の処理をスキップするようにすればよい。
【０１２９】
そのようにすれば、原画像であるカラー画像Ｐの対応する画素ｐijのＲＧＢ成分が、そのまま、画像Ｑ’の対応する画素ｑij’のＲＧＢ成分となるので、電子透かしは画像Ｑ’から完全に除去されることになる。
【０１３０】
Ａ−３．第１の実施例の効果：
以上説明したように、本実施例によれば、サーバコンピュータ１００からクライアントコンピュータ２００に画像データを配信し、その後は、画像購入を条件として、状態変更情報を配信するだけで、クライアントコンピュータ２００において、画像に埋め込まれていた電子透かしを見えなくすることができ、サンプルであった画像を、不都合なく利用できる画像に容易に変換することができる。従って、サーバ側（配信者側）においては、原画像のデータとは別にサンプル用のデータを用意する必要がない。また、クライアント側（視聴者側）においては、同じ画像のデータを２度ダウンロードする手間をとる必要がない。
【０１３１】
また、電子透かしを見えなくするために、可視状態にある電子透かしを不可視状態に変更する方法を採った場合は、画像から電子透かしが見えなくなった後であっても、電子透かしは不可視状態で埋め込まれたままとなる。従って、画像が不正利用された場合でも、埋め込まれている電子透かしを検出することにより、不正利用の発見や追跡を容易に行うことができる。
【０１３２】
Ｂ．第２の実施例：
Ｂ−１．情報配信システムの構成：
図１１は本発明の第２の実施例としての情報配信システムの構成を示す説明図である。第１の実施例では、電子透かしの埋め込まれた画像データと、電子透かしを除去または不可視状態に変更するための状態変更情報と、を同じサーバコンピュータ１００から配信していたが、本実施例では、異なるサーバコンピュータから配信するようにしている。従って、本実施例の情報配信システムは、画像データの配信を行うサーバコンピュータ４００と、状態変更情報の配信を行うサーバコンピュータ５００と、クライアントコンピュータ２００と、課金管理コンピュータ３５０と、これらコンピュータが接続されるネットワーク３００と、で構成されている。
【０１３３】
このうち、サーバコンピュータ４００及び５００は、基本的には、第１の実施例で述べたサーバコンピュータ１００と同様の構成となっているが、メインメモリ４０４，５０４に格納されるコンピュータプログラムが異なっている。
【０１３４】
即ち、サーバコンピュータ４００におけるメインメモリ４０４には、画像データに電子透かしを可視状態で埋め込む電子透かし埋め込み部４０４ａと、電子透かしの埋め込まれた画像データを送信する画像データ送信部４０４ｂの、各機能を実現するためのコンピュータプログラムが格納されているのに対し、サーバコンピュータ５００におけるメインメモリ５０４には、可視状態にある電子透かしを除去するか不可視状態に変更するための状態変更情報を送信する状態変更情報送信部５０４ａの機能を実現するためのコンピュータプログラムが格納されている。従って、それ以外の構成は、概ね、サーバコンピュータ１００と同様であるので、詳しい説明は省略する。
【０１３５】
Ｂ−２．情報配信処理の内容：
図１１の情報配信システムが行う情報配信処理は、基本的には、第１の実施例における情報配信処理と同じであるが、画像データはサーバコンピュータ４００から、状態変更情報はサーバコンピュータ５００から、別々に、クライアントコンピュータ２００に配信する点が異なっている
具体的に、図２に示したフローチャートに従って説明すると、まず、サーバコンピュータ４００において、電子透かし埋め込み部４０４ａが、配信すべき対象となる画像データに、電子透かしを可視状態で埋め込む処理を行う（ステップＳ１２）。次に、サーバコンピュータ４００のＣＰＵ４０２が、その画像の存在をネットワーク３００を介してクライアントコンピュータ２００に知らせる（ステップＳ１４）。その知らせに応答して、クライアントコンピュータ２００のＣＰＵ２０２が、画像のダウンロードをサーバコンピュータ４００に対して要求する（ステップＳ１６）と、サーバコンピュータ４００では、画像データ送信部４０４ｂが、電子透かしが可視状態で埋め込まれた画像データを、ネットワーク３００を介してクライアントコンピュータ２００に送信する（ステップＳ１８）。
【０１３６】
その後、クライアントコンピュータ２００のユーザが、画像の購入を決定し、クライアントコンピュータ２００と課金管理コンピュータ３５０との間で課金処理（ステップＳ２２，Ｓ２４）が行われると、課金管理コンピュータ３５０は、そのユーザに対する課金完了の情報をネットワーク３００を介して、今度は、サーバコンピュータ５００に送る（ステップＳ２６）。サーバコンピュータ５００では、その情報を受け取ると、クライアントコンピュータ２００のユーザが画像を購入したことを認識し、状態変更情報送信部５０４ａが、可視状態にある電子透かしを不可視状態にするための状態変更情報を、ネットワーク３００を介してクライアントコンピュータ２００に送信する（ステップＳ２８）。このとき、状態変更情報送信部５０４ａは、秘密鍵である状態変更情報を暗号化した上で、クライアントコンピュータ２００に送信する。
【０１３７】
それ以降の処理は、第１の実施例の場合と同様であるので、説明は省略する。なお、本実施例においては、サーバコンピュータ４００において、画像データに対する電子透かしの埋め込みを行っているので、状態変更情報の元になる、変換行列Ｂにおける明るさ係数ｚと透かし係数ｘの各値の情報は、当初、サーバコンピュータ４００内に存在している。従って、もう一方のサーバコンピュータ５００からクライアントコンピュータ２００に状態変更情報を配信するためには、それら状態変更情報の元になる情報と、それに対応する画像に関する情報とを、サーバコンピュータ４００からサーバコンピュータ５００に伝送する必要がある。具体的な伝送方法としては、それらの情報を暗号化した上で、ネットワーク３００を介して伝送するか、または、それらの情報を専用線を介して直接伝送するかの方法がとられる。サーバコンピュータ５００では、そのようにして受け取った状態変更情報や画像に関する情報を対応付けた上で管理している。
【０１３８】
Ｂ−３．第２の実施例の効果：
以上説明したように、本実施例においても、サーバコンピュータ４００からクライアントコンピュータ２００に画像のデータを配信し、その後は、画像購入を条件として、サーバコンピュータ５００からクライアントコンピュータ２００に状態変更情報を配信するだけで、クライアントコンピュータ２００において、画像に埋め込まれていた電子透かしを見えなくすることができ、サンプルであった画像を、不都合なく利用できる画像に容易に変換することができる。従って、サーバ側（配信者側）においては、原画像のデータとは別にサンプル用のデータを用意する必要もなく、クライアント側（視聴者側）においては、同じ画像を２度ダウンロードする手間をとる必要がないという効果を奏することができる。
【０１３９】
また、本実施例においては、サーバコンピュータ４００は、画像データを専用に扱って、電子透かしの埋め込みや画像データの送信のみを担当し、サーバコンピュータ５００は、状態変更情報のみを扱って、状態変更情報の管理や送信のみを担当しているので、各々のサーバコンピュータの負荷が軽くて済む。また、画像データを扱うサーバコンピュータ４００の負荷は、状態変更情報のみを扱うサーバコンピュータ５００に比べて重くなるので、ネットワーク３００上に、画像データを扱うサーバコンピュータ４００を多数用意し、それらから得られる状態変更情報を１台のサーバコンピュータ５００によって管理するようにしてもよい。
【０１４０】
Ｃ．第３の実施例：
Ｃ−１．情報配信システムの構成：
図１２は本発明の第３の実施例としての情報配信システムの構成を示す説明図である。第１及び第２の実施例では、電子透かしの埋め込まれた画像データをサーバコンピュータから配信するようにしていたが、本実施例では、そのような画像データをＣＤ−ＲＯＭに記録し、そのＣＤ−ＲＯＭをクライアントコンピュータ２００のユーザに配布するようにしている。従って、本実施例の情報配信システムでは、画像データの配信を行うサーバコンピュータは不要となり、状態変更情報の配信を行うサーバコンピュータ５００と、ＣＤ−ＲＯＭに記録された画像データの読み取りが可能であると共に、配信された状態変更情報を受信することが可能なクライアントコンピュータ６００と、課金管理コンピュータ３５０と、これらコンピュータが接続されるネットワーク３００と、で構成されている。
【０１４１】
このうち、クライアントコンピュータ６００は、基本的には、第１の実施例で述べたクライアントコンピュータ２００と同様の構成となっているが、メインメモリ６０４に格納されるコンピュータプログラムが異なっている。具体的には、クライアントコンピュータ６００のメインメモリ６０４には、サーバコンピュータ５００から送信された状態変更情報を受信する状態変更情報受信部６０４ａと、状態変更情報に基づいて、画像データに可視状態で埋め込まれている電子透かしを除去するか不可視状態に変更する状態変更部６０４ｂの、各機能を実現するためのコンピュータプログラムが格納されている。従って、それ以外の構成は、概ね、クライアントコンピュータ２００と同様であるので、詳しい説明は省略する。
【０１４２】
Ｃ−２．情報配信処理の内容：
本実施例では、前提として、前述したとおり、画像データはＣＤ−ＲＯＭに記録して、クライアントコンピュータ６００のユーザに配布される。その画像データには予め電子透かしが可視状態で埋め込まれている。ＣＤ−ＲＯＭの配布の方法としては、そのＣＤ−ＲＯＭを書籍などに添付して配布するようにしてもよいし、店頭や街頭で直接配布するようにしてもよい。あるいは、特定のユーザだけに郵送や宅配便にて配送するようにしてもよい。
【０１４３】
こうして、クライアントコンピュータ６００のユーザが、画像データの記録されたＣＤ−ＲＯＭを手に入れると、ユーザは、そのＣＤ−ＲＯＭをＣＤ−ＲＯＭ６１８と同じようにして、ＣＤ−ＲＯＭドライブ６１６に装着し、記録された画像データの読み取りと表示を指示する。すると、クライアントコンピュータ６００のＣＰＵ６０２は、ＣＤ−ＲＯＭドライブ６１６によって、その画像データをＣＤ−ＲＯＭから読み出して、表示装置６１２に表示させる。
【０１４４】
このとき表示される画像には、前述したとおり、電子透かしが可視状態で埋め込まれているため、電子透かしが画像の邪魔になって、このままでは、この画像を利用するのは不都合な状態となっている。すなわち、ＣＤ−ＲＯＭに記録されている画像は、ユーザに対してサンプル用として提供されていることになる。
【０１４５】
図１３は図１２の情報配信システムが行う情報配信処理の手順を示すフローチャートである。この処理は、状態変更情報送信部５０４ａ、状態変更情報受信部６０４ａ、状態変更部６０４ｂなどの処理として実現されている。
【０１４６】
ユーザが、表示した画像を気に入り、この画像を不都合なく利用するために、この画像の購入を決定し、その旨をクライアントコンピュータ６００に指示すると、図１３に示す情報配信処理が起動される。まず、クライアントコンピュータ６００は、画像購入の指示をネットワーク３００を介して課金管理コンピュータ３５０に送る（ステップＳ４２）。課金管理コンピュータ３５０では、課金管理情報にアクセスして、クライアントコンピュータ２００のユーザを特定し、そのユーザに対して画像購入に対する課金を行う（ステップＳ４４）。このとき、そのユーザが新規ユーザである場合には、課金管理コンピュータ３５０は、ユーザ登録手続完了後に、上記した課金処理を行う。
【０１４７】
次に、課金管理コンピュータ３５０は、そのユーザに対する課金完了の情報をネットワーク３００を介してサーバコンピュータ５００に送る（ステップＳ４６）。サーバコンピュータ５００では、その情報を受け取ると、クライアントコンピュータ２００のユーザが、配布した画像を購入したことを認識する。
【０１４８】
サーバコンピュータ５００では、画像購入を認識したら、状態変更情報送信部５０４ａが、可視状態にある電子透かしを除去または不可視状態に変更するための状態変更情報を、ネットワーク３００を介してクライアントコンピュータ６００に送信する（ステップＳ４８）。このとき、状態変更情報送信部１０４ｃは、秘密鍵である状態変更情報を暗号化した上で、クライアントコンピュータ６００に送信する。
【０１４９】
クライアントコンピュータ６００では、状態変更情報受信部６０４ａが、送信された状態変更情報を受信して復号化し、元の状態変更情報を得る（ステップＳ５０）。その後、状態変更部６０４ｂが、ＣＤ−ＲＯＭドライブ６１６によって、電子透かしが埋め込まれている対象となる画像データをＣＤ−ＲＯＭから読み出し、取得した状態変更情報に基づいて、その電子透かしを可視状態から不可視状態にする処理を行い（ステップＳ５２）、それによって得られた画像データをハードディスク６１０に格納する。
【０１５０】
以上の手順を踏むことによって、配布されたＣＤ−ＲＯＭに記録された画像データであっても、埋め込まれている電子透かしを可視状態から不可視状態に変更して、画像を不都合なく利用できるようにすることが可能となる。
【０１５１】
なお、本実施例においては、画像データに予め電子透かしを可視状態で埋め込み、その画像データをＣＤ−ＲＯＭに記録した上で、ユーザに配布している。従って、画像データに対する電子透かしの埋め込みは、基本的に、サーバコンピュータ５００とは別の場所で行われている。そのため、状態変更情報の元になる、変換行列Ｂにおける明るさ係数ｚと透かし係数ｘの各値の情報を、埋め込み処理の行われている場所から取得して、サーバコンピュータ５００に入力する必要がある。入力に当たっては、状態変更情報だけでなく、それに対応する画像に関する情報も入力する。サーバコンピュータ５００では、そのようにして入力された状態変更情報や画像に関する情報を対応付けた上で管理する。
【０１５２】
Ｃ−３．第３の実施例の効果：
以上説明したように、本実施例では、クライアントコンピュータ６００において、画像データの記録されているＣＤ−ＲＯＭの配布を受けた場合に、画像購入を条件として、サーバコンピュータ５００からクライアントコンピュータ２００に状態変更情報の配信を受けるだけで、画像に埋め込まれていた電子透かしを見えなくすることができ、サンプルであった画像を、不都合なく利用できる画像に容易に変換することができる。従って、サーバ側（配信者側）においては、原画像のデータとは別にサンプル用のデータを用意する必要もなく、クライアント側（視聴者側）においては、原画像を手に入れるために、画像データをダウンロードする手間をとる必要がない。さらにまた、可視状態にある電子透かしを不可視状態に変更する方法を採った場合は、画像から電子透かしが見えなくなった後であっても、電子透かしは不可視状態で埋め込まれたままである。よって、その画像が不正利用された場合でも、その電子透かしを検出することにより、不正利用の発見や追跡を容易に行うことができる。
【０１５３】
Ｄ．第４の実施例：
Ｄ−１．情報配信システムの構成：
図１４は本発明の第４の実施例としての情報配信システムの構成を示す説明図である。第１の実施例では、サーバコンピュータ１００から画像データを配信していたが、本実施例では、サーバコンピュータから音声データを配信するようにしている。従って、本実施例の情報配信システムは、音声データなどの種々の情報を配信するサーバコンピュータ７００と、その音声データなどを受信し、再生することが可能なクライアントコンピュータ８００と、課金管理コンピュータ３５０と、これらのコンピュータが接続されるネットワーク３００と、で構成されている。
【０１５４】
このうち、サーバコンピュータ７００は、基本的には、第１の実施例で述べたサーバコンピュータ１００と、クライアントコンピュータ８００はクライアントコンピュータ２００と、それぞれ同様の構成となっているが、メインメモリ７０４，８０４に格納されるコンピュータプログラムが異なっている。
【０１５５】
即ち、サーバコンピュータ７００におけるメインメモリ７０４には、対象となる音声データに電子透かしを可聴状態で埋め込む電子透かし埋め込み部７０４ａと、電子透かしの埋め込まれた音声データを送信する音声データ送信部７０４ｂと、可聴状態にある電子透かしを除去するか不可聴状態にするための状態変更情報を送信する状態変更情報送信部７０４ｃの、各機能を実現するためのコンピュータプログラムが格納されている。電子透かし埋め込み部７０４ａ、音声データ送信部７０４ｂ及び状態変更情報送信部７０４ｃの機能については後で詳しく説明する。
【０１５６】
一方、クライアントコンピュータ８００におけるメインメモリ８０４には、サーバコンピュータから送信された音声データを受信する音声データ受信部８０４ａと、サーバコンピュータから送信された状態変更情報を受信する状態変更情報受信部８０４ｂと、音声データに可聴状態で埋め込まれている電子透かしを、状態変更情報に基づいて除去するか不可聴状態に変更する状態変更部８０４ｃの、各機能を実現するためのコンピュータプログラムが格納されている。音声データ受信部８０４ａ、状態変更情報受信部８０４ｂ及び状態変更部８０４ｃの機能については後で詳しく説明する。
【０１５７】
このように、本実施例においては、対象となるデータが音声データであるので、それに埋め込まれる電子透かしは、知覚可能な状態において、聴覚によって認識できる情報である。例えば、音や、人などの声や、楽曲などの他、雑音などが含まれる。なお、本明細書において、電子透かしが可聴状態にあるとは、聴覚によって認識できる状態にあることを言い、不可聴状態にあるとは、聴覚による認識が困難である状態を言う。
【０１５８】
Ｄ−２．情報配信処理の内容：
図１５は図１４の情報配信システムが行う情報配信処理の手順を示すフローチャートである。この処理は、電子透かし埋め込み部７０４ａ、音声データ送信部７０４ｂ、状態変更情報送信部７０４ｃ、音声データ受信部８０４ａ、状態変更情報受信部８０４ｂ、状態変更部８０４ｃなどの処理として実現されている。
【０１５９】
ところで、上述した第１の実施例では、画像データに電子透かしを埋め込むために、変換行列Ｂを利用していたが、本実施例では、音声データに電子透かしを埋め込むために、離散コサイン変換（ＤＣＴ変換）を利用するようにしている。ここで、ＤＣＴ変換について簡単に説明する。
【０１６０】
ＤＣＴ変換は、信号を周波数成分に変換する直交変換の一手法である。離散コサイン変換（ＤＣＴ変換）及び逆変換（ＩＤＣＴ変換）は、以下の式（８）〜（１０）で表される。
【０１６１】
【数８】

【０１６２】
【数９】

【０１６３】
【数１０】

【０１６４】
ここで、ｇ（ｎ）は入力データ、Ｇ（ｍ）は変換係数であり、また、変換対象となる入力データブロックのサイズはＮである。
【０１６５】
ＤＣＴ変換は、一般に２次元の画像を対象として、動画像圧縮の国際標準であるＭＰＥＧや静止画像圧縮の国際標準であるＪＰＥＧに使われており、画像ごとに大きく違う画素値の分布を、画像依存性の少ない一定の性質をもつ変換係数に変換する性質をもっている。ＤＣＴ変換はこのような性質を有しているので、変換係数に透かし情報を埋め込むと、これを逆変換して得られる画像においては、透かし情報が画像全体に拡散することになる。
【０１６６】
従って、このＤＣＴ変換を１次元の音声に適用しても、同様に、変換係数に透かし情報を埋め込み、これを逆変換して音声を得ると、その音声においては、透かし情報が音声全体に拡散することになる。
【０１６７】
Ｎの大きさを有するブロックに対する変換係数は、１個の直流成分Ｇ（０）と、Ｎ²−１個の交流成分とで構成される。また、直流成分の係数値が最大となり、それに近い交流成分ほどその係数値は大きくなる傾向にある。すなわち、画像を形成するエネルギーの大部分が低周波に集中する。このとき、各ブロックに対する変換係数値の分布は、直流成分をピークにもつラプラス分布により近似できる。このような変換係数に、透かし情報を自然な形で埋め込むためには、このようなラプラス分布を大きく変化させないように透かし情報を埋め込めばよい。
【０１６８】
ところで、透かし情報をＤＣＴ変換すれば、そのエネルギの大部分は低周波に集中し、ラプラス分布により近似できる分布を有するはずである。そこで、原音声の変換係数と、透かし情報の変換係数とにそれぞれ所定の比を乗じて加算すれば、自然なかたちで透かし情報を変換係数に埋め込むことが可能である。また、この変換係数を空間領域に逆変換すると、透かし情報が音声全体に分散されることになる。従って、周波数領域と空間領域のいずれにおいても、第三者が、電子透かしの埋め込まれている音声から電子透かしを不正に除去することは困難である。
【０１６９】
さて、それでは、図１５に戻って、図１５に示す情報配信処理が起動されると、まず、サーバコンピュータ７００において、電子透かし埋め込み部７０４ａが、配信すべき対象となる音声データに、電子透かしを可聴状態で埋め込む処理を行う（ステップＳ６２）。
【０１７０】
具体的には、電子透かし埋め込み部７０４ａは、図１６に示す埋め込み処理の手順に従って、電子透かしの埋め込みを行う。図１６は図１の電子透かし埋め込み部１０４ａが行う埋め込み処理の手順を示す説明図である。電子透かし埋め込み部１０４ａは、まず、電子透かしを埋め込むべき対象となる音声（原音声）Ｓと、埋め込むべき電子透かしである透かし音声Ｔと、を特定する処理を行なう。これらの処理は、ハードディスク７１０に記憶された音声ファイルを指定することにより行なわれる。
【０１７１】
なお、以下の議論のために、原音声Ｓに電子透かしである透かし音声Ｔを埋め込んだ後の音声（以下、署名済み音声と呼ぶ）をＵとする。
【０１７２】
両音声Ｓ，Ｔを用意した後、次に、原音声ＳをＤＣＴ変換することによって、第１のＤＣＴ変換係数Ｘｓを得る。また、透かし音声Ｔを同様にＤＣＴ変換することによって、第２のＤＣＴ変換係数Ｘｔを得る。
【０１７３】
次に、式（１１）に従って、第１と第２のＤＣＴ変換係数の対応する周波数成分同士を合成し、第３のＤＣＴ変換係数Ｘｕを生成する。
【０１７４】
【数１１】

【０１７５】
すなわち、対応する各周波数成分ごとに、第１のＤＣＴ係数Ｘｓに重み係数ａ／ｎを乗じ、第２のＤＣＴ係数Ｘｔに重み係数ｂ／ｍを乗じて加算（または減算）することによって、第３のＤＣＴ変換係数Ｘｕが生成される。なお、２つの重み係数ａ／ｎ，ｂ／ｍは、それぞれ０でない任意の値である。２つの重み係数ａ／ｎ，ｂ／ｍは、有理数であること（すなわち、ａ，ｂ，ｎ，ｍが整数であること）が好ましく、また、その和を１に設定することが好ましい。これらの２つの重み係数ａ／ｎ，ｂ／ｍは、原音声Ｓと透かし音声ＴのＤＣＴ係数を合成する際の相対比を表すものと考えることも可能である。
【０１７６】
第３のＤＣＴ係数Ｘｕには透かし情報が含まれているので、「透かし付きＤＣＴ係数」とも呼ぶ。この透かし付きＤＣＴ係数Ｘｕを逆変換（ＩＤＣＴ）すると、署名済み音声Ｕが得られる。すなわち、この署名済み音声Ｕは、原音声Ｓに電子透かしである透かし音声Ｔが可聴状態で埋め込まれた音声である。従って、この署名済み音声Ｕのデータを再生すると、原音声Ｓと共に、透かし音声Ｔが重なり合って聞こえてくる。
【０１７７】
以上のようにして、署名済み音声、即ち、電子透かしが可聴状態で埋め込まれた音声が得られたら、次に、サーバコンピュータ７００のＣＰＵ７０２は、その音声の存在をＷｅｂページや電子メールなどによりネットワーク３００を介してクライアントコンピュータ８００に知らせる（ステップＳ６４）。その知らせを受けて、クライアントコンピュータ８００のユーザが、その音声のダウンロードを希望すれば、そのユーザの指示によって、クライアントコンピュータ８００のＣＰＵ８０２は、その音声のダウンロードをサーバコンピュータ７００に対して要求する（ステップＳ６６）。
【０１７８】
サーバコンピュータ７００では、その要求を受け取ると、音声データ送信部７０４ｂが、電子透かしが可聴状態で埋め込まれた音声データをハードディスク７１０から読み出して、ネットワーク３００を介してクライアントコンピュータ８００に送信する（ステップＳ６８）。クライアントコンピュータ８００では、音声データ受信部８０４ａが、サーバコンピュータ７００から送信された音声データを受信して、ハードディスク８１０に格納する（ステップＳ７０）。
【０１７９】
それから、クライアントコンピュータ８００のユーザが、その音声の再生を指示すると、クライアントコンピュータ８００のＣＰＵ８０２は、その音声データをハードディスク８１０から読み出して、音声出力装置８１４で出力させる。このとき、再生される音声には、電子透かしが可聴状態で埋め込まれているため、前述したとおり、ユーザには、原音声の他に、透かし音声が重なり合って聞こえてくる。従って、原音声を聞くためには、透かし音声が邪魔になり、この音声を利用するのには不都合となっている。つまり、この状態では、この音声は、ユーザに対してサンプル用として提供されていることになる。
【０１８０】
従って、ユーザが、この音声を気に入った場合には、可聴状態で埋め込まれている電子透かしを聞こえなくするために、この音声を購入する必要がある。
【０１８１】
そこで、ユーザが、この音声の購入を決定し、その旨をクライアントコンピュータ８００に指示すると、クライアントコンピュータ８００は、音声購入の指示をネットワーク３００を介して課金管理コンピュータ３５０に送り（ステップＳ７２）、課金管理コンピュータ３５０では、その指示を受け取ると、クライアントコンピュータ８００のユーザを特定し、そのユーザに対して、音声購入に対する課金を行う（ステップＳ７４）。
【０１８２】
そして、課金管理コンピュータ３５０は、そのユーザに対する課金完了の情報をネットワーク３００を介してサーバコンピュータ７００に送る（ステップＳ７６）。サーバコンピュータ７００では、その情報を受け取ると、クライアントコンピュータ８００のユーザが、配信した音声を購入したことを認識する。
【０１８３】
ところで、可聴状態で埋め込んだ電子透かしを聞こえなくする場合、電子透かしを音声データから除去してしまう方法と、電子透かしを音声データ内に留めたまま、聴覚では認識し難い状態、即ち、不可聴状態に変更する方法とが考えられる。前者の方法を用いた場合には、音声データ内にもはや電子透かしが残らないので、その音声が不正利用された場合に、その発見や追跡が困難になる。よって、後者の方法を採った方が、不正利用の防止や著作権保護の観点からも望ましい。
【０１８４】
従って、本実施例では、電子透かしを可視状態から不可視状態に変更する後者の方法について、まず説明する。サーバコンピュータ７００では、音声購入を認識したら、状態変更情報送信部７０４ｃが、可聴状態にある電子透かしを不可聴状態にするための状態変更情報を、ネットワーク３００を介してクライアントコンピュータ８００に送信する（ステップＳ７８）。
【０１８５】
状態変更情報とは、具体的には、式（１１）で示した第２のＤＣＴ係数Ｘｔと、重み係数ｂ／ｍに近似した係数ｂ／ｍ−Δの各値である。これらの値は、電子透かしを埋め込んだ者のみが知る秘密鍵であり、この値を用いることによって、可聴状態にある電子透かしを不可聴状態に変更することが可能となる。従って、状態変更情報送信部７０４ｃは、この状態変更情報を第三者に知られないようにするために、暗号化した上で、クライアントコンピュータ８００に送信する。
【０１８６】
なお、状態変更情報の伝送には、サーバコンピュータ７００からの電子メールを利用してもよいし、クライアントコンピュータ８００からのダウンロードによって行なうようにしてもよい。
【０１８７】
クライアントコンピュータ８００では、状態変更情報受信部８０４ｂが、送信されきた状態変更情報を受信し、暗号化されているその状態変更情報を復号化する（ステップＳ８０）。そして、状態変更部８０４ｃが、ハードディスク８１０から電子透かしの埋め込まれている音声データを読み出し、復号化した状態変更情報に基づいて、その電子透かしを可聴状態から不可聴状態に変更する処理を行う（ステップＳ８２）。
【０１８８】
具体的には、状態変更部８０４ｃは、図１７に示す状態変更処理の手順に従って、電子透かしの状態変更を行う。図１７は図１４の状態変更部８０４ｃが行う状態変更処理の手順を示す説明図である。状態変更部８０４ｃは、まず、受信した状態変更情報から、第２のＤＣＴ係数Ｘｔと、重み係数ｂ／ｍに近似した係数ｂ／ｍ−Δの各値を特定すると共に、電子透かしの埋め込まれた音声、すなわち、署名済み音声Ｕを特定する。署名済み音声Ｕの特定は、ハードディスク８１０に記憶された音声ファイルを指定することにより行なわれる。
【０１８９】
次に、状態変更部８０４ｃは、特定された署名済み音声ＵをＤＣＴ変換することによって、第３のＤＣＴ変換係数Ｘｕを得る。そして、その第３のＤＣＴ変換係数Ｘｕに対して、先に特定した第２のＤＣＴ係数Ｘｔと係数ｂ／ｍ−Δの各値を用いて、式（１２）に示す演算を行って、第４のＤＣＴ係数Ｘｕ’を得る。
【０１９０】
【数１２】

【０１９１】
この演算によって、第４のＤＣＴ係数Ｘｕ’には、第３のＤＣＴ変換係数Ｘｕに含まれていた透かし情報分±（ｂ／ｍ）Ｘｔは取り除かれるものの、新たにごく少量の±ΔＸｔ分の透かし情報が含まれるようになる。従って、この第４のＤＣＴ係数Ｘｕ’を逆変換（ＩＤＣＴ）して、署名済み音声Ｕ’を生成すると、その署名済み音声Ｕ’には透かし音声Ｔがごく少量しか埋め込まれていないので、この署名済み音声Ｕ’のデータを再生しても、ほとんど原音声Ｓしか聞こえてこず、透かし音声Ｔを聴覚によって認識することは困難である。すなわち、この署名済み音声Ｕ’は、原音声Ｓに電子透かしである透かし音声Ｔが不可聴状態で埋め込まれた音声となる。
【０１９２】
こうして、可聴状態にある電子透かしを不可聴状態に変更させる状態変更処理が終了すると、その後、クライアントコンピュータ８００のユーザが、音声の再生を指示した場合、クライアントコンピュータ８００のＣＰＵ８０２は、電子透かしが不可聴状態となった音声データをハードディスク８１０から読み出し、音声出力装置８１４に出力させる。このとき再生される音声を聞くと、電子透かしがあたかも除去されたように聞こえ、従って、ユーザは、この音声を不都合なく利用することが可能となる。
【０１９３】
但し、電子透かしは不可聴状態になっただけで、音声から完全に除去されたわけではないので、仮に、この音声が不正利用された場合でも、その発見や追跡を容易に行うことができる。
【０１９４】
即ち、この音声が不正利用されたという疑いがある時には、原音声Ｓを用いて、この原音声Ｓと署名済み音声Ｕ’との差分をとることにより、不可聴状態で埋め込まれていた電子透かし、即ち、透かし音声Ｔを、痕跡が十分確認できる程度に抽出することができる。
【０１９５】
ところで、前述したように、可聴状態で埋め込んだ電子透かしを聞こえなくする方法としては、電子透かしを音声データから除去してしまう方法と、電子透かしを不可聴状態に変更する方法と、があり、上記の説明では、後者の方法を用いる場合について説明したが、前者の方法を用いる場合には、状態変更情報として、式（１１）で示した第２のＤＣＴ係数Ｘｔと重み係数ｂ／ｍを得ると共に、状態変更処理として、次のような処理を行うようにする。
【０１９６】
具体的には、状態変更部８０４ｃは、図１８に示す状態変更処理の手順に従って、電子透かしの状態変更を行う。図１８は図１４の状態変更部８０４ｃが行う他の状態変更処理の手順を示す説明図である。状態変更部８０４ｃは、まず、受信した状態変更情報から、第２のＤＣＴ係数Ｘｔと重み係数ｂ／ｍの各値を特定すると共に、ハードディスク８１０に記憶された音声ファイルから、電子透かしの埋め込まれた音声、すなわち、署名済み音声Ｕを特定する。
【０１９７】
次に、状態変更部８０４ｃは、特定された署名済み音声ＵをＤＣＴ変換することによって、第３のＤＣＴ変換係数Ｘｕを得て、さらに、その第３のＤＣＴ変換係数Ｘｕに対して、第２のＤＣＴ係数Ｘｔと係数ｂ／ｍの各値を用いて、式（１３）に示す演算を行って、第５のＤＣＴ係数Ｘｓ’を得る。
【０１９８】
【数１３】

【０１９９】
この演算によって、第５のＤＣＴ係数Ｘｕ’では、第３のＤＣＴ変換係数Ｘｕに含まれていた透かし情報分±（ｂ／ｍ）Ｘｔが完全に取り除かれる。従って、この第５のＤＣＴ係数Ｘｓ’を逆変換（ＩＤＣＴ）して、音声Ｓ’を生成すると、その音声Ｓ’には透かし音声Ｔが全く埋め込まれていないので、この音声をＳ’のデータを再生すると、原音声Ｓしか聞こえない。この音声Ｓ’は、可聴状態で埋め込まれていた電子透かしが完全に除去された音声となる。
【０２００】
Ｄ−３．第４の実施例の効果：
以上説明したように、本実施例によれば、サーバコンピュータ７００からクライアントコンピュータ８００に音声データを配信し、その後に、音声購入を条件として、状態変更情報を配信するだけで、クライアントコンピュータ８００において、音声に可聴状態で埋め込まれていた電子透かしを聞こえなくすることができ、サンプルであった音声を、不都合なく利用できる音声に容易に変換することができる。従って、サーバ側（配信者側）においては、原音声のデータとは別にサンプル用のデータを用意する必要がない。また、クライアント側（視聴者側）においては、同じ音声のデータを２度ダウンロードする手間をとる必要がない。
【０２０１】
また、電子透かしを聞こえなくするために、可聴状態にある電子透かしを不可聴状態に変更する方法を採った場合は、音声から電子透かしが聞こえなくなった後であっても、電子透かしは不可聴状態で埋め込まれたままとなるので、その後、その音声が不正利用された場合でも、埋め込まれている電子透かしを検出することにより、不正利用の発見や追跡を容易に行うことができる。
【０２０２】
Ｅ．変形例：
なお、この発明は上記の実施例や実施形態に限られるものではなく、その要旨を逸脱しない範囲において種々の態様において実施することが可能であり、例えば次のような変形も可能である。
【０２０３】
Ｅ−１．変形例１：
上記した第４の実施例においては、音声データを対象とし、サーバコンピュータ７００からは、電子透かしの埋め込まれた音声データと、電子透かしを除去または不可聴状態に変更するための状態変更情報と、を配信するようにしていた。しかしながら、音声データを対象とする場合、この実施例に限定されるものではなく、画像データを対象とした場合における第２の実施例と同様に、電子透かしの埋め込まれた音声データと、電子透かしを除去または不可視状態に変更するための状態変更情報と、を異なるサーバコンピュータから配信するようにしてもよく、また、第３の実施例と同様に、電子透かしの埋め込まれた音声データをＣＤ−ＲＯＭに記録し、そのＣＤ−ＲＯＭをクライアントコンピュータのユーザに配布するようにしてもよい。
【０２０４】
Ｅ−２．変形例２：
上記した第１の実施例においては、電子透かしとなる透かし画像として、図６に示したようなロゴマーク画像を用いる場合を例として説明したが、本発明はこれに限定されるものではない。
【０２０５】
例えば、電子透かしとなる透かし画像や透かし音声として、商品やサービス、会社や店舗などに関する広告・宣伝を載せた画像や音声を用いるようにすれば、新たな効果が期待できる。すなわち、画像データや音声データの提供者が、スポンサをとって、そのスポンサに関する広告・宣伝を載せた画像や音声を、それら画像データや音声データに知覚可能な状態で埋め込んで、視聴者に配信または配布することにより、それら画像データや音声データ自体が新たな広告・宣伝媒体となり得る。また、スポンサから提供者に広告・宣伝費が支払われることにより、電子透かしを除去または知覚困難な状態にするための状態変更情報を視聴者に対して無料で配信することも可能となる。
【０２０６】
Ｅ−３．変形例３：
上記した第１の実施例においては、電子透かしの埋め込みの対象となる画像として、また、電子透かしとなる透かし画像として、図５及び図６に示したように、何れも、静止画像を用いる場合を例として説明したが、本発明は、対象画像，透かし画像の何れも、静止画像に限らず、動画像を用いるようにしてもよい。
【０２０７】
また、上記した説明では、対象データとして画像データを用いる実施例（即ち、第１ないし第３の実施例）と、音声データを用いる実施例（即ち、第４の実施例）と、を分けて説明したが、対象データとして、例えば、映画などのような画像データ，音声データを共に用いる場合についても、本発明は、当然に適用可能である。
【０２０８】
Ｅ−４．変形例４：
上記した第１の実施例においては、画像データに電子透かしを式（２）に示した変換行列Ｂを利用して埋め込むようにしたが、画像データに対する電子透かしの埋め込み方法は、これに限定されるものではなく、種々の方法を用いることができる。
【０２０９】
例えば、本願の出願人によって既に出願されている特願平１１−１７０８１６号に記載されているように、ＲＧＢ表色系の画像データを式（１４）に示す変換行列Ａを用いてＹＣｂＣｒ表色系の画像データに変換する際に、式（１５）に示す変換行列Ｘを用いて透かし画像を埋め込み、式（１４）に示す逆変換行列Ａ^-1を用いて逆変換することによって、署名済み画像を得るようにしてもよい。
【０２１０】
【数１４】

【０２１１】
【数１５】

【０２１２】
但し、ｘは、前述した透かし係数である。
【０２１３】
また、同じく、本願の出願人によって既に出願されている特願平１１−２９６６４８号に記載されているように、画像データを、４つの画素ずつ、式（１６）に示す変換行列Ｃを用いて変換することにより、画像データに電子透かしを埋め込むようにしてもよい。
【０２１４】
【数１６】

【０２１５】
但し、ｄは各画素の輝度成分に影響を与える明るさ係数であり、ｂ１，ｂ２，ｂ３は透かし情報を生成するための透かし係数である。
【０２１６】
また、上記した第４の実施例では、音声データに電子透かしを埋め込むために、ＤＣＴ変換を利用したが、画像データに電子透かしを埋め込む場合にも、ＤＣＴ変換を利用するようにしてもよい。この場合、本願の出願人によって既に出願されている特願平１１−２９６２０４号に記載されているように、ＤＣＴ変換として２次元のＤＣＴ変換を用いるようにする。
【０２１７】
Ｅ−５．変形例５：
上記した実施例においては、サーバコンピュータがクライアントコンピュータに対して、状態変更情報を配信する条件として、課金管理コンピュータ３５０が課金処理を行い、サーバコンピュータが、課金管理コンピュータ３５０からの課金完了の情報を受け取って、クライアントコンピュータのユーザが配信または配布した画像や音声を購入したことを認識するという条件を用いているが、本発明はこれに限定されるものではなく、種々の条件を用いることができる。
【０２１８】
例えば、クライアントコンピュータからサーバコンピュータに対して、ユーザＩＤと所定のパスワードが送信されたことを条件としてもよいし、クライアントコンピュータのユーザがサーバコンピュータの管理者に対し、電話やファックスなどによって、購入の意志を示したことを条件としてもよい。
【０２１９】
Ｅ−６．変形例６：
上記した第３の実施例では、電子透かしの埋め込まれた画像データをＣＤ−ＲＯＭに記録して配布するようにしたが、本発明はこれに限定されるものではなく、電子透かしの埋め込まれた画像データや音声データを、フレキシブルディスクや、ハードディスクや、ＤＶＤ−ＲＯＭや、光磁気ディスクや、ＩＣカードや、ＲＯＭカートリッジや、磁気テープカセットなどに記録して配布するようにしてもよい。
【０２２０】
Ｅ−７．変形例７：
上記した第４の実施例においては、第３のＤＣＴ変換係数Ｘｕを生成する際に、式（１１）に従って、第１と第２のＤＣＴ変換係数の対応する周波数成分同士を合成するようにしていたが、このように対応する周波数同士で合成させる必要は必ずしもなく、異なる周波数同士で合成させるようにしてもよい。その場合には、配信される状態変更情報に、どの周波数同士を合成させたかの情報を含ませる必要がある。
また、上記した第４の実施例においては、ＤＣＴ変換を用いるようにしていたが、ＦＦＴやウェーブレット変換など、他の種類の直交変換を採用することも可能である。
【０２２１】
Ｅ−８．変形例８：
上記した第１の実施例では、サーバコンピュータ１００において、第２の実施例では、サーバコンピュータ４００において、第４の実施例では、サーバコンピュータ７００において、それぞれ、画像データまたは音声データに電子透かしを埋め込むようにしていたが、本発明はこれに限定されるものではなく、電子透かしの埋め込みは、上記サーバコンピュータ以外のところで行ってもよい。すなわち、上記サーバコンピュータは、そのような電子透かしの埋め込まれた画像データや音声データを格納し、クライアントコンピュータに対して配信することが可能であればよい。
【図面の簡単な説明】
【図１】本発明の第１の実施例としての情報配信システムの構成を示す説明図である。
【図２】図１の情報配信システムが行う情報配信処理の手順を示すフローチャートである。
【図３】図１の電子透かし埋め込み部１０４ａが行う埋め込み処理の手順を示すフローチャートである。
【図４】埋め込みの対象となるカラー画像Ｐにおける画素ＰijのＲＧＢ成分を説明するための説明図である。
【図５】埋め込み対象となるカラー画像Ｐの一例を示す説明図である。
【図６】埋め込むべき透かし画像Ｗの一例を示す説明図である。
【図７】図５のカラー画像Ｐに図６の透かし画像Ｗを埋め込んで得られる署名済み画像の一例を示す説明図である。
【図８】図１の状態変更部２０４ｃが行う状態変更処理の手順を示すフローチャートである。
【図９】図７の署名済み画像について可視のマーク部分を不可視化して得られる新たな署名済み画像の一例を示す説明図である。
【図１０】図９の署名済み画像から不可視状態にあるマーク部分を抽出して得られるロゴマーク画像の一例を示す説明図である。
【図１１】本発明の第２の実施例としての情報配信システムの構成を示す説明図である。
【図１２】本発明の第３の実施例としての情報配信システムの構成を示す説明図である。
【図１３】図１２の情報配信システムが行う情報配信処理の手順を示すフローチャートである。
【図１４】本発明の第４の実施例としての情報配信システムの構成を示す説明図である。
【図１５】図１４の情報配信システムが行う情報配信処理の手順を示すフローチャートである。
【図１６】図１の電子透かし埋め込み部１０４ａが行う埋め込み処理の手順を示す説明図である。
【図１７】図１４の状態変更部８０４ｃが行う状態変更処理の手順を示す説明図である。
【図１８】図１４の状態変更部８０４ｃが行う他の状態変更処理の手順を示す説明図である。
【符号の説明】
１００…サーバコンピュータ
１０２…ＣＰＵ
１０４…メインメモリ
１０４ａ…電子透かし埋め込み部
１０４ｂ…画像データ送信部
１０４ｃ…状態変更情報送信部
１０６…キーボード
１０８…マウス
１１０…ハードディスク
１１２…表示装置
１１４…音声出力装置
１１６，２１６…ＣＤ−ＲＯＭドライブ
１１６…ＣＤ−ＲＯＭドライブ
１１８…ＣＤ−ＲＯＭ
１２０…通信装置
１２２…バス
２００…クライアントコンピュータ
２０２…ＣＰＵ
２０４…メインメモリ
２０４ａ…画像データ受信部
２０４ｂ…状態変更情報受信部
２０４ｃ…状態変更部
２０６…キーボード
２０８…マウス
２１０…ハードディスク
２１２…表示装置
２１４…音声出力装置
２１６…ＣＤ−ＲＯＭドライブ
２１８…ＣＤ−ＲＯＭ
２２０…通信装置
２２２…バス
３００…ネットワーク
３５０…課金管理コンピュータ
４００…サーバコンピュータ
４０２…ＣＰＵ
４０４…メインメモリ
４０４ａ…電子透かし埋め込み部
４０４ｂ…画像データ送信部
４０６…キーボード
４０８…マウス
４１０…ハードディスク
４１２…表示装置
４１４…音声出力装置
４１６…ＣＤ−ＲＯＭドライブ
４１８…ＣＤ−ＲＯＭ
４２０…通信装置
４２２…バス
５００…サーバコンピュータ
５０４…メインメモリ
５０４ａ…状態変更情報送信部
５０６…キーボード
５０８…マウス
５１０…ハードディスク
５１２…表示装置
５１４…音声出力装置
５１６…ＣＤ−ＲＯＭドライブ
５１８…ＣＤ−ＲＯＭ
５２０…通信装置
５２２…バス
６００…クライアントコンピュータ
６０２…ＣＰＵ
６０４…メインメモリ
６０４ａ…状態変更情報受信部
６０４ｂ…状態変更部
６０６…キーボード
６０８…マウス
６１０…ハードディスク
６１２…表示装置
６１４…音声出力装置
６１６…ＣＤ−ＲＯＭドライブ
６１８…ＣＤ−ＲＯＭ
６２０…通信装置
６２２…バス
７００…サーバコンピュータ
７０２…ＣＰＵ
７０４…メインメモリ
７０４ａ…電子透かし埋め込み部
７０４ｂ…音声データ送信部
７０４ｃ…状態変更情報送信部
７０６…キーボード
７０８…マウス
７１０…ハードディスク
７１２…表示装置
７１４…音声出力装置
７１６…ＣＤ−ＲＯＭドライブ
７１８…ＣＤ−ＲＯＭ
７２０…通信装置
７２２…バス
８００…クライアントコンピュータ
８０２…ＣＰＵ
８０４…メインメモリ
８０４ａ…音声データ受信部
８０４ｂ…状態変更情報受信部
８０４ｃ…状態変更部
８０６…キーボード
８０８…マウス
８１０…ハードディスク
８１２…表示装置
８１４…音声出力装置
８１６…ＣＤ−ＲＯＭドライブ
８１８…ＣＤ−ＲＯＭ
８２０…通信装置
８２２…バス[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for distributing audio data such as images and music such as still images and moving images.
[0002]
[Prior art]
In recent years, with the development of network technologies such as the Internet, technologies for distributing various information via a network have been proposed. For example, it is possible to sample a still image, a moving image, or a musical piece at a predetermined sampling period, convert it into a digital signal, and distribute this via a network. Viewers can get the latest still images, videos and music while staying at home.
[0003]
In the conventional distribution system, when still images, videos and music are distributed for a fee, the thumbnail images and a part of the music are prepared in advance as sample data, and the viewer uses the sample data. Download from the distribution server via the network, watch the sample, and if you like it, pay the fee and retrieve the original still image, video, or song data from the distribution server again via the network Things have been done.
[0004]
In another distribution system, focusing on the low cost of manufacturing a CD-ROM, the above-described sample data is recorded on a CD-ROM for still images, moving images, and music, and this CD-ROM is viewed by the viewer. It is also distributed to. In this case, the viewer who received the distribution sees and listens to the sample recorded on the CD-ROM, and pays a fee if he / she likes it. You will get video and music data.
[0005]
[Problems to be solved by the invention]
However, each of the above-described conventional distribution systems has a problem that sample data must be prepared separately from the data finally distributed. In this case, if the sample is music, it may be data that includes music for several tens of seconds below the intro, or data that has been degraded in sound quality by reducing the sampling frequency. It had to be prepared separately from the data.
[0006]
From the viewer side, in the former distribution system, once the sample data is downloaded from the distribution server via the network, and the music is liked and purchased, it is again transmitted from the distribution server via the network. There is a problem in that it is necessary to download the data of the original music, and the user has to download the same music twice. Even in the latter distribution system, it is necessary to download data from the distribution server via the network in order to obtain the original music despite the distribution of the CD-ROM. There was a problem that it took time to download data.
[0007]
The above-mentioned problem is the same not only in the case of music but also in the case where still images and moving images are distributed for a fee.
[0008]
Accordingly, an object of the present invention is to provide an information distribution system capable of solving the above-described problems of the prior art and obtaining an original image or sound from data once distributed or distributed as a sample or the like. is there.
[0009]
[Means for solving the problems and their functions and effects]
In order to achieve at least a part of the above object, a first information distribution system of the present invention is an information distribution system for distributing information from a server computer to a client computer,
The server computer
A data transmission unit that transmits data representing at least one of an image and audio and embedded with the digital watermark in a perceptible state to the client computer;
A state change information transmitting unit that transmits state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied;
With
The client computer is
A data receiving unit for receiving the target data transmitted from the server computer;
A state change information receiving unit for receiving the state change information transmitted from the server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
It is a summary to provide.
[0010]
As described above, in the first information distribution system of the present invention, the data transmission unit of the server computer is data representing at least one of the image and the sound, and the digital watermark is embedded in a perceivable state. When the target data is transmitted, the data receiving unit of the client computer receives the target data. Thereafter, in the server computer, when a predetermined condition is satisfied, the state change information transmission unit transmits state change information for removing a digital watermark in a perceptible state or making it difficult to perceive. The state change information receiving unit of the client computer receives the state change information. Then, in the client computer, the state change unit removes or changes the digital watermark embedded in the perceptible state in the target data to a state that is difficult to perceive based on the state change information.
[0011]
Therefore, according to the first information distribution system of the present invention, target data representing an image or sound is distributed from a server computer to a client computer, and then state change information is distributed when a predetermined condition is satisfied. The client computer can remove the digital watermark embedded in the perceived state of the target data or change it to a state that is difficult to perceive, and convert the sampled data into data that can be used without inconvenience. Can be easily converted. Therefore, it is not necessary for the server side (distributor side) to prepare sample data separately from the original data. Further, it is not necessary for the client side (viewer side) to download the same target data twice.
[0012]
Further, when a digital watermark that is in a perceptible state is changed to a state in which it is difficult to perceive, for example, the digital watermark remains embedded in the target data even in a state in which it is difficult to perceive. Therefore, even when the target data is illegally used, it is possible to easily detect and track illegal use by detecting the embedded digital watermark.
[0013]
The second information distribution system of the present invention is an information distribution system for distributing information from the first and second server computers to the client computer,
The first server computer is
A data transmitting unit that transmits data representing at least one of an image and sound and embedded in a state in which a digital watermark is perceptible to the client computer;
The second server computer is
A state change information transmission unit configured to transmit state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied; ,
The client computer is
A data receiving unit for receiving the target data transmitted from the first server computer;
A state change information receiving unit for receiving the state change information transmitted from the second server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
It is a summary to provide.
[0014]
As described above, in the second information distribution system of the present invention, the data transmission unit of the first server computer is data representing at least one of an image and a sound, and a digital watermark can be perceived. When the embedded target data is transmitted, the data receiving unit of the client computer receives the target data. Thereafter, in a second server computer different from the first server computer when the predetermined condition is satisfied, the state change information transmitting unit removes the digital watermark in a perceptible state or is in a state where it is difficult to perceive State change information is transmitted, and the state change information receiving unit of the client computer receives the state change information. Then, in the client computer, the state change unit removes or changes the digital watermark embedded in the perceptible state in the target data to a state that is difficult to perceive based on the state change information.
[0015]
Therefore, according to the second information distribution system of the present invention, the same effect as that of the first information distribution system described above can be obtained, and the first server computer handles the target data exclusively, Since the second server computer handles only the status change information and only transmits the status change information, the load on each server computer is light. I'll do it.
[0016]
A third information distribution system according to the present invention provides information from a server computer to a client computer having data representing at least one of an image and sound and having target data embedded with a digital watermark perceivable. An information distribution system for distributing
The server computer
A state change information transmission unit configured to transmit state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied; ,
The client computer is
A state change information receiving unit for receiving the state change information transmitted from the server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
It is a summary to provide.
[0017]
As described above, in the third information distribution system of the present invention, the client computer is data that represents at least one of an image and sound and is embedded in a state in which a digital watermark can be perceived. I have data. In the server computer, when a predetermined condition is satisfied, the state change information transmitting unit transmits state change information for removing a digital watermark in a perceptible state or making it difficult to perceive. The state change information receiving unit of the client computer receives the state change information. Then, in the client computer, the state change unit removes or changes the digital watermark embedded in the perceptible state in the target data to a state that is difficult to perceive based on the state change information.
[0018]
Therefore, according to the third information distribution system of the present invention, when the client computer has data representing an image or sound, the state change information is distributed from the server computer to the client computer by satisfying a predetermined condition. The client computer can remove the digital watermark embedded in the perceived state of the target data or change it to a state that is difficult to perceive, and convert the sampled data into data that can be used without inconvenience. Can be easily converted. Therefore, on the server side (distributor side), it is not necessary to prepare sample data separately from the original data. On the client side (viewer side), in order to obtain the original data, There is no need to download it.
[0019]
In addition, when a digital watermark in a perceptible state is changed to a state in which it is difficult to perceive, for example, the digital watermark remains embedded in the target data even in a state in which it is difficult to perceive. Therefore, even when the target data is illegally used, it is possible to easily detect and track illegal use by detecting the embedded digital watermark.
[0020]
In the third information distribution system of the present invention, the target data of the client computer may be recorded on a recording medium readable by the client computer.
[0021]
This is because target data representing an image or sound may be distributed to viewers after being recorded on a recording medium.
[0022]
In the information distribution system of the present invention, it is preferable that the digital watermark includes at least one of advertisement information and advertisement information.
[0023]
As described above, since the digital watermark includes advertisement information and advertisement information, the digital watermark not only functions to clearly indicate the copyright and prevent illegal use, but also serves as an advertisement and advertisement. The target data embedded with it can be a new advertisement / promotion medium.
[0024]
In the information distribution system of the present invention, it is preferable that the predetermined condition is that the server computer recognizes that an image or sound represented by the target data has been purchased by a user of the client computer.
[0025]
If the provided image or sound is not purchased by the user but the status change information is distributed to the user, the user can freely use the image and sound. This is because the voice provider may suffer a great loss.
[0026]
In the information distribution system of the present invention, the state change information transmission unit encrypts and transmits the state change information, and the state change information reception unit receives and decrypts the encrypted state change information. It is preferable to do.
[0027]
This is because if the state change information is known to a third party, the digital watermark may be removed or made difficult to perceive by the person, and the provided image or sound may be used illegally.
[0028]
In the information distribution system of the present invention, when the target data represents at least an image, the digital watermark is information that can be visually recognized when the digital watermark is in a perceptible state, and the digital watermark is difficult to perceive. When it is in a state, it is preferable that visual recognition becomes difficult.
[0029]
In this way, when the target data is image data recognized visually, the digital watermark is also information that can be visually recognized in a perceptible state and difficult to recognize visually in a difficult state. There must be.
[0030]
In the information distribution system of the present invention, when the target data represents at least sound, the digital watermark is information that can be perceived by hearing when the digital watermark is in a perceptible state, and the digital watermark is difficult to perceive. When it is in a state, it is preferable that recognition by hearing is difficult.
[0031]
In this way, when the target data is audio data recognized by auditory sense, the digital watermark is also information that can be recognized by auditory sense in a perceptible state and difficult to perceive in a perceptible state. There must be.
[0032]
A first server computer of the present invention is a server computer that distributes information to a client computer,
A data transmission unit that transmits data representing at least one of an image and audio and embedded with the digital watermark in a perceptible state to the client computer;
A state change information transmitting unit that transmits state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied;
It is a summary to provide.
[0033]
As described above, in the first server computer of the present invention, the data transmission unit is the data representing at least one of the image and the sound, and the target data embedded in a state where the digital watermark can be perceived, A state for transmitting to the client computer, and when the predetermined condition is satisfied, the state change information transmitting unit removes the digital watermark in a perceptible state or makes it difficult to perceive in the client computer Send change information.
[0034]
Therefore, according to the first server computer of the present invention, when the target data representing the image or the sound is distributed from the server computer and then the predetermined condition is satisfied, the state change information is only distributed. In a computer, digital watermarks embedded in the perceived state of the target data can be removed or changed to a state that is difficult to perceive, and the sampled data can be easily converted to data that can be used without inconvenience. be able to. Therefore, it is not necessary for the server side (distributor side) to prepare sample data separately from the original data.
[0035]
A first client computer of the present invention is a client computer that receives information distributed from a server computer,
A data receiving unit that receives target data that is transmitted from the first server computer and that represents at least one of an image and sound and in which a digital watermark is perceptible;
Status change for removing the perceptible digital watermark transmitted from the first server computer or a second server computer different from the first server computer or making it difficult to perceive A state change information receiver for receiving information;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
It is a summary to provide.
[0036]
As described above, in the first client computer of the present invention, the data receiving unit is data representing at least one of the image and the sound transmitted from the first server computer, and the digital watermark can be perceived. The perceptible state received from the first server computer or the second server computer different from the first server computer while receiving the target data embedded in the state The state change information for removing the digital watermark in the state or making it difficult to perceive is received. Then, the state change unit removes or changes the digital watermark embedded in the perceptible state in the target data to a state in which it is difficult to perceive based on the received state change information.
[0037]
Therefore, according to the first client computer of the present invention, the client computer can perceive the target data only by receiving the target data representing the image or sound distributed from the server computer and then receiving the state change information. The digital watermark embedded in the state can be removed or changed to a state that is difficult to perceive, and the sampled data can be easily converted into data that can be used without any inconvenience. Therefore, it is not necessary for the client side (viewer side) to download the same target data twice.
[0038]
A second server computer of the present invention is a server computer that distributes information to a client computer,
The gist of the present invention is to include a data transmission unit that transmits data representing at least one of an image and audio and embedded with the digital watermark perceivable to the client computer.
[0039]
In this way, the server computer handles the target data exclusively and is in charge of only transmitting the target data in which the digital watermark is embedded, so that the load on the server computer can be reduced.
[0040]
A third server computer of the present invention is a server computer that distributes information to a client computer having target data embedded in a state in which a digital watermark is perceptible, which is data representing at least one of an image and sound Because
A state change information transmitting unit configured to transmit state change information for removing the digital watermark in a perceptible state or making it difficult to perceive when a predetermined condition is satisfied; This is the gist.
[0041]
As described above, in the third server computer of the present invention, when a predetermined condition is satisfied, the state change information transmission unit is in a state that can be perceived by the client computer having the target data. The state change information for removing the digital watermark or making it difficult to perceive is transmitted.
[0042]
Therefore, according to the third server computer of the present invention, when the client computer has data representing an image or sound, the client computer only has to deliver the state change information from the server computer by satisfying a predetermined condition. , The digital watermark embedded in the target data in a perceptible state can be removed or changed to a state that is difficult to perceive, and the sampled data can be easily converted into data that can be used without any inconvenience. Can do. Therefore, on the server side (distributor side), it is not necessary to prepare sample data separately from the original data.
[0043]
The second client computer of the present invention is data representing at least one of image and sound, and has target data embedded with a digital watermark perceivable, and information distributed from the server computer A client computer that receives
A state change information receiving unit that receives state change information transmitted from the server computer for removing the digital watermark in a perceptible state or making it difficult to perceive;
Based on the state change information, a state change unit that changes the digital watermark embedded in the data to be recorded from a perceptible state to a difficult perceptible state;
It is a summary to provide.
[0044]
As described above, the second client computer of the present invention has data representing at least one of an image and a sound in advance and having target data embedded in a state where a digital watermark can be perceived. Yes. Then, the state change receiving unit receives state change information transmitted from the server computer for removing the perceptible digital watermark or making it difficult to perceive, and the state change unit receives the target data The digital watermark embedded in a perceptible state is removed or changed to a state that is difficult to perceive based on the state change information.
[0045]
Therefore, according to the second client computer of the present invention, when data representing an image or sound is included, the client computer can perceive the target data only by receiving the status change information from the server computer. The digital watermark embedded in (1) can be removed or changed to a state that is difficult to perceive, and the sampled data can be easily converted into data that can be used without any inconvenience. Accordingly, it is not necessary for the client side (viewer side) to download the data in order to obtain the original data.
[0046]
The present invention relates to an information distribution system, a server computer and a client computer constituting the information distribution method, an information distribution method, a computer program for realizing the apparatus or method, a recording medium recording the computer program, and a computer program Including a data signal embodied in a carrier wave, and the like.
[0047]
Other aspects of the invention
The present invention can also adopt other aspects as described below. A first aspect is an information distribution method for distributing information to a client computer,
(A) transmitting data representing at least one of an image and a sound, the target data embedded in a state where a digital watermark can be perceived, to the client computer;
(B) transmitting a state change information for removing the digital watermark in a perceptible state or making it difficult to perceive when a predetermined condition is satisfied;
An information distribution method comprising:
[0048]
A second aspect is an information receiving method for receiving information distributed from a server computer,
(A) receiving the target data that is transmitted from the first server computer and represents at least one of the image and the sound and embedded with the digital watermark being perceivable;
(B) To remove the digital watermark in a perceptible state transmitted from the first server computer or a second server computer different from the first server computer, or to make it difficult to perceive Receiving state change information;
(C) removing the digital watermark embedded in the target data in a perceptible state based on the state change information or changing the digital watermark to a state that is difficult to perceive;
An information receiving method comprising:
[0049]
A third aspect is an information delivery method for delivering information to a client computer having target data embedded with data representing at least one of an electronic image and audio and a digital watermark perceivable. There,
(A) An information distribution method comprising a step of transmitting state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer.
[0050]
A fourth aspect is an information receiving method for receiving information distributed from a server computer,
(A) preparing data that represents at least one of an image and sound and embedded with a digital watermark being perceivable;
(B) receiving state change information transmitted from the server computer for removing the digital watermark in a perceptible state or making it difficult to perceive;
(C) based on the state change information, changing the digital watermark embedded in the data to be recorded from a perceptible state to a difficult perceptible state;
An information receiving method comprising:
[0051]
The effects of the present invention described above can also be achieved by adopting such various aspects.
[0052]
DETAILED DESCRIPTION OF THE INVENTION
Next, embodiments of the present invention will be described in the following order based on examples.
A. First embodiment:
A-1. Information distribution system configuration:
A-2. Information distribution process:
A-3. Effects of the first embodiment:
B. Second embodiment:
B-1. Information distribution system configuration:
B-2. Information distribution process:
B-3. Effects of the second embodiment:
C. Third embodiment:
C-1. Information distribution system configuration:
C-2. Information distribution process:
C-3. Effects of the third embodiment:
D. Fourth embodiment:
D-1. Information distribution system configuration:
D-2. Information distribution process:
D-3. Effects of the fourth embodiment:
E. Variations:
E-1. Modification 1:
E-2. Modification 2:
E-3. Modification 3:
E-4. Modification 4:
E-5. Modification 5:
E-6. Modification 6:
E-7. Modification 7:
E-8. Modification 8:
[0053]
A. First embodiment:
A-1. Information distribution system configuration:
FIG. 1 is an explanatory diagram showing the configuration of an information distribution system as a first embodiment of the present invention. This information distribution system includes a server computer 100, a client computer 200, a charge management computer 350, and a network 300 to which these computers are connected.
[0054]
Among these, the server computer 100 is a computer for distributing various information such as image data to the client computer 200. The server computer 100 includes a CPU 102, a main memory 104 that is an internal storage device including a ROM and a RAM, a keyboard 106, a mouse 108, a hard disk 110 that is an external storage device, a display device 112 including a CRT, and the like. An audio output device 114 including a speaker, a CD-ROM drive 116, a communication device 120 including a network card and a modem, and a bus 122 for connecting these components are provided. In FIG. 1, various interface circuits are omitted. The communication device 120 is connected to the network 300 via a communication line.
[0055]
The main memory 104 includes a digital watermark embedding unit 104a that embeds a digital watermark in the target image data in a visible state, an image data transmission unit 104b that transmits the image data in which the digital watermark is embedded, and a digital watermark that is in a visible state. A computer program for realizing each function of the state change information transmitting unit 104c that transmits state change information for removing or changing to an invisible state is stored. The functions of the digital watermark embedding unit 104a, the image data transmission unit 104b, and the state change information transmission unit 104c will be described in detail later.
[0056]
Thus, in this embodiment, the target data is image data, and the digital watermark embedded in the data is information that can be visually recognized in a perceptible state. For example, characters, figures, shapes, patterns, colors, still images and moving images are included. Note that in this specification, the electronic watermark is in a visible state means that the electronic watermark is in a visually recognizable state, and the invisible state means a state in which visual recognition is difficult.
[0057]
On the other hand, the client computer 200 is a computer for receiving various information such as image data distributed from the server computer 100. As with the server computer 100, the client computer 200 includes a CPU 202, a main memory 204 that is an internal storage device including a ROM and a RAM, a keyboard 206, a mouse 208, a hard disk 210 that is an external storage device, a CRT, and the like. A display device 212, a sound output device 214 such as a speaker, a CD-ROM drive 216, a communication device 220 such as a network card or a modem, and a bus 222 for connecting these components. . In FIG. 1, various interface circuits are omitted. The communication device 220 is also connected to the network 300 via a communication line.
[0058]
The main memory 204 includes an image data receiving unit 204a that receives image data transmitted from the server computer, a state change information receiving unit 204b that receives state change information transmitted from the server computer, and the state change information. The computer program for realizing each function of the state changing unit 204c for removing the digital watermark embedded in the image data in the visible state or changing it to the invisible state is stored. The functions of the image data receiving unit 204a, the state change information receiving unit 204b, and the state changing unit 204c will be described in detail later.
[0059]
Computer programs that realize the functions of the units 104a to 104c and 204a to 204c stored in the main memory 104 of the server computer 100 and the main memory 204 of the client computer 200 are computer-readable recording media such as CD-ROMs 118 and 218. Provided in recorded form. The computer reads the computer program from the recording medium using the CD-ROM drives 116 and 216 and stores the computer program directly or in an external storage device such as the hard disk 110 and then transfers it to an internal storage device such as the main memory 104 or 204. . Alternatively, a program supply device (not shown) may be provided on the network 300 so that the computer program is supplied to the computer via a communication line. When realizing the function of the computer program, the computer program stored in the internal storage device is executed by a microprocessor such as the CPUs 102 and 202 of the computer. Further, the computer program recorded on the recording medium may be read by the computer and directly executed.
[0060]
The “recording medium” in the present invention includes a flexible disk, a CD-ROM, a DVD-ROM, a magneto-optical disk, an IC card, a ROM cartridge, a punch card, a printed matter on which a code such as a barcode is printed, and the inside of a computer. A variety of computer-readable media such as a storage device (memory such as RAM and ROM) and an external storage device can be used.
[0061]
In this specification, the computer is a concept including a hardware device and an operation system, and means a hardware device that operates under the control of the operation system. Further, when an operation system is not required and a hardware device is operated by an application program or firmware alone, the hardware device itself corresponds to a computer. The hardware device includes at least a microprocessor such as a CPU and means for reading a computer program recorded on a recording medium. For example, even when a CPU, a ROM, or the like is incorporated in an electronic device such as a mobile phone, a portable information terminal, a player, a recorder, or the like and the electronic device has a function as a computer, Naturally included in the concept of computers. The computer program includes program code for causing such a computer to realize the functions of the above-described means. Note that some of the functions described above may be realized by an operation system instead of an application program. Furthermore, a program for embedding a digital watermark, a process for transmitting or receiving an image or sound, a process for receiving, or a process for changing the state of a digital watermark is added to the program for performing an image or sound process in the form of a plug-in. It is good also as a thing.
[0062]
In this embodiment, in the client computer 200, dedicated browser software is used as a computer program for realizing the functions of the image data receiving unit 204a, the state change information receiving unit 204b, and the state changing unit 204c.
[0063]
In FIG. 1, a billing management computer 350 performs billing management, bills the user of the client computer 200 in accordance with an image purchase instruction from the client computer 200, and transmits the billing completion information to the server computer 100. It is. This billing management computer 350 is also connected to the network 300 via a communication line by a communication device or the like.
[0064]
A-2. Information distribution process:
FIG. 2 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG. This processing is realized as processing of the digital watermark embedding unit 104a, the image data transmission unit 104b, the state change information transmission unit 104c, the image data reception unit 204a, the state change information reception unit 204b, the state change unit 204c, and the like.
[0065]
As shown in the figure, when this information distribution process is started, first, in the server computer 100, the digital watermark embedding unit 104a performs a process of embedding the digital watermark in the image data to be distributed in a visible state ( Step S12).
[0066]
Specifically, the digital watermark embedding unit 104a embeds a digital watermark according to the procedure of the embedding process shown in FIG. FIG. 3 is a flowchart showing the procedure of the embedding process performed by the digital watermark embedding unit 104a of FIG. First, the digital watermark embedding unit 104a performs processing for specifying an image (color image) P to be embedded with a digital watermark and a watermark image W that is a digital watermark to be embedded (step S105). These processes are performed by designating an image file stored in the hard disk 110.
[0067]
In the present embodiment, the target color image P is represented by an RGB (red, green, blue) color system, is composed of m × n pixels, and each pixel has a color of 8 bits (that is, A color image represented by tristimulus values R, G, and B (hereinafter referred to as RGB components) having 256 gradations) was used. here,
P = {pij | i = 0, 1, 2,... M-1; j = 0, 1, 2,.
It shall be expressed as
[0068]
Further, as the watermark image W, a binary image having m × n pixels having the same size as that of the color image P and the color of each pixel being black or white was used. Specifically, as described later, a logo mark image with “LOGO1” written in black characters on a white background was prepared. here,
W = {wij | wij = 1, 0: i = 0, 1, 2,... M-1; j = 0, 1, 2,.
It shall be expressed as That is, the pixel value is 1 for the mark portion (black portion) and 0 for the non-mark portion (white portion).
[0069]
However, the watermark image W is not necessarily required to have the same size as the color image P.
[0070]
For the following discussion, let Q denote an image (hereinafter referred to as a signed image) after embedding a watermark image W, which is a digital watermark, in the color image P. This signed image Q is also composed of m × n pixels having the same size as the color image P and the watermark image W. So this is
Q = {qij | i = 0, 1, 2,... M-1; j = 0, 1, 2,.
It shall be expressed as
[0071]
After preparing both images P and W, next, variables i and j for image processing are set to an initial value 0 (steps S110 and S115). Then, for the variables i and j, it is determined whether or not the value of the pixel wij of the watermark image W is 1 (that is, a mark portion) (step S120).
[0072]
As a result of the determination, if the value of the pixel wij is 0 (that is, the non-marked portion), the pixel pij of the color image P corresponding to the pixel wij is selected, and the RGB component of the pixel pij is directly used as the signed image. The RGB component of the pixel qij corresponding to Q is set (step S130).
[0073]
That is, if the RGB component of the pixel pij in the color image P is (rij, gij, bij) and this is represented by a three-dimensional vector, as shown in FIG. 4, Pij = [rij, gij, bij] ^t It becomes. Further, when the RGB component of the pixel qij in the signed image Q is (Rij, Gij, Bij) and expressed as a three-dimensional vector, qij = [Rij, Gij, Bij] ^t It becomes. However, the subscript t represents transposition. Therefore, when the value of the pixel wij is 0, it is expressed as in Expression (1).
[0074]
[Expression 1]

[0075]
On the other hand, when the value of the pixel wij is 1 (that is, the mark portion), the pixel pij of the color image P corresponding to the pixel wij is selected, and the RGB component of the pixel pij is converted by the equation (2). The RGB component obtained by conversion by the matrix B is used as the RGB component of the corresponding pixel qij of the signed image Q (step S125).
[0076]
[Expression 2]

[0077]
That is, when the value of the pixel wij is 1, it is expressed as in Expression (3).
[0078]
[Equation 3]

[0079]
As shown in equations (2) and (3), the transformation matrix B is a cubic square matrix, and all of the diagonal components are composed of the value z (where z is an arbitrary real number), and the other components. Consists of the value x (where x is any real number other than 0) or -x. The sum of the components in each row and each column is z. Among these, the value z, which is a diagonal component, has a significant influence on the luminance component of each pixel, and therefore in the present specification, this value z is referred to as a brightness coefficient. The remaining coefficient x is a secret coefficient for generating watermark information, and in the present specification, this value x is called a watermark coefficient. A color difference signal to be a watermark signal is obtained based on the calculation result of the watermark coefficient.
[0080]
However, as described above, in this embodiment, the brightness coefficient z in the transformation matrix B is set to a value other than 1 in order to embed the watermark image W, that is, the digital watermark, in the visible state. Then, based on Expression (3), the RGB component of the mark portion pixel in the signed image Q is expressed as Expression (4).
[0081]
[Expression 4]

[0082]
Looking at the right side of equation (4), in the first term, the RGB components of the pixels of the mark portion in the color image P are respectively multiplied by the brightness coefficient z, and the brightness components of the three primary colors are multiplied by this brightness coefficient z. Will be equally affected. In the second term, a color difference signal which is a watermark signal is newly added. As a result, the watermark information is embedded in a visible state based on the brightness coefficient z, and the watermark information based on the watermark coefficient x is potentially embedded.
[0083]
Here, if the brightness coefficient z is set to a positive real value, the mark portion can be displayed in black in the range of 0 <z <1, and the mark portion can be displayed in white in the range of z> 1. It is possible to display. At this time, as the value of the brightness coefficient z is closer to 1, the brightness of the mark portion becomes lighter and approaches an invisible state. On the other hand, as the value of the brightness coefficient z departs from 1, a clearer luminance is obtained at the mark portion.
[0084]
Note that when embedding a watermark image in a visible state, it is easy to cause the pixel value to overflow beyond the upper limit value or to overflow under the lower limit value by greatly changing the luminance. There must be. If such an overflow or underflow occurs, the pixel value is forcibly replaced with an upper limit value or a lower limit value to be corrected.
[0085]
If the obtained RGB component is not an integer as a result of conversion by the conversion matrix B, the decimal part is rounded off to an integer.
[0086]
After performing the arithmetic processing as described above, the variables i and j are sequentially incremented (steps S135 and S145), and it is determined whether or not the processing has been completed for all the pixels of the watermark image W (steps S140 and S150). Until the process is completed, the above process of embedding the watermark image W, which is a digital watermark, into the color image P is repeated. When the process for all pixels is completed, the watermark image embedding process is terminated.
[0087]
Here, a specific example when the watermark image W, which is a digital watermark, is embedded in the color image P in a visible state will be described.
[0088]
In this specific example, the portrait image shown in FIG. 5A and the cafeteria image shown in FIG. 5B are used as the target color image P. These images are obtained by reducing color digital standard image data issued by the Japanese Standards Association, and are 320 × 256 pixels in size and RGB components are 8 bits each. On the other hand, the logo mark image shown in FIG. 6 was used as the watermark image W that is an electronic watermark to be embedded.
[0089]
When embedding the visible logo mark image shown in FIG. 6 in the portrait image shown in FIG. 5A, in order to adopt a black display, the brightness coefficient z is set to 0.7, The watermark coefficient x is set to 0.1. However, when embedding in the cafeteria image shown in FIG. 5B, the brightness coefficient z is changed so that the logo mark can be recognized more clearly by changing the brightness of the logo mark from that in the portrait image. Was set to 0.4. The watermark coefficient x was set to the same 0.1.
[0090]
As a result of such embedding, each signed image is represented as shown in FIGS. As is apparent from a comparison between FIGS. 7A and 7B, the visible logo mark can be displayed with different brightness depending on the value of the brightness coefficient z.
[0091]
As described above, when a signed image, that is, an image in which a digital watermark is embedded in a visible state, is obtained, the CPU 102 of the server computer 100 determines the presence of the image, such as a Web page or an e-mail. Using the means, the client computer 200 is notified via the network 300 (step S14). In response to the notification, if the user of the client computer 200 desires to acquire the image, the CPU 202 of the client computer 200 requests the server computer 100 to download the image in accordance with the user's instruction (step S100). S16).
[0092]
When the server computer 100 receives the request, the image data transmission unit 104b reads the image data in which the digital watermark is embedded in a visible state from the hard disk 110, and transmits the image data from the communication device 120 to the client computer 200 via the network 300. (Step S18). In the client computer 200, the image data receiving unit 204a receives the image data transmitted from the server computer 100 via the communication device 220 and stores it in the hard disk 210 (step S20).
[0093]
Thereafter, when the user of the client computer 200 instructs display of the image, the CPU 202 of the client computer 200 reads out the image data from the hard disk 210 and causes the display device 212 to display the image data. As a result, the user can grasp the contents of the image sufficiently, but since the digital watermark is embedded in the visible state in the image displayed at this time, as shown in FIGS. 7A and 7B, The digital watermark interferes with the image, and it is inconvenient to use this image as it is. That is, in this state, this image is provided for the user as a sample.
[0094]
Therefore, if the user likes this image and wishes to use this image without any inconvenience, it is necessary to purchase this image in order to hide the digital watermark embedded in the visible state. .
[0095]
When the user decides to purchase this image and instructs the client computer 200 to do so, the client computer 200 sends an image purchase instruction to the charge management computer 350 via the network 300 (step S22). Upon receipt of the instruction, billing management computer 350 accesses the billing management information inside, specifies the user of client computer 200, and charges the user for image purchase (step S24).
[0096]
At this time, if the user is a new user, billing management computer 350 instructs client computer 600 to perform a user registration procedure prior to billing processing. Then, if it is confirmed that the user registration procedure has been performed by the user of the client computer 600, the charging management computer 350 performs the above-described charging process.
[0097]
Then, billing management computer 350 sends billing completion information for the user to server computer 100 via network 300 (step S26). When receiving the information, the server computer 100 recognizes that the user of the client computer 200 has purchased the distributed image.
[0098]
By the way, when the watermark embedded in the visible state is made invisible, the method of removing the digital watermark from the image data and the state where the digital watermark remains in the image data and is difficult to visually recognize, that is, the invisible state. And a method of changing to In the former method, since the digital watermark is completely lost from the image data, it becomes difficult to find and track the image when it is illegally used. Therefore, the latter method is desirable from the viewpoint of preventing unauthorized use.
[0099]
Therefore, in the present embodiment, the latter method for changing the digital watermark embedded in the visible state to the invisible state will be described first.
[0100]
Therefore, first, when the server computer 100 recognizes the purchase of the image, the state change information transmission unit 104c sends the state change information for making the visible digital watermark invisible to the client computer 200 via the network 300. Transmit (step S28).
[0101]
Specifically, the state change information is each value of the brightness coefficient z and the watermark coefficient x in the transformation matrix B used in step S125 described above. These values are secret keys that are known only to the person who embeds the digital watermark. By using these values, it is possible to change the visible digital watermark to the invisible state.
[0102]
Therefore, the state change information transmission unit 104c encrypts and transmits the state change information, which is a secret key, to the client computer 200 so that the third party does not know the state change information.
[0103]
The state change information may be transmitted by using an e-mail from the server computer 100 or by downloading from the client computer 200.
[0104]
In the client computer 200, the state change information receiving unit 204b receives the transmitted state change information, and decrypts the encrypted state change information (step S30). Then, the state changing unit 204c reads the image data in which the digital watermark is embedded from the hard disk 210, and performs a process of changing the digital watermark from the visible state to the invisible state based on the decoded state change information (step S1). S32).
[0105]
Specifically, the state changing unit 204c changes the state of the digital watermark according to the procedure of the state changing process shown in FIG. FIG. 8 is a flowchart showing the procedure of the state change process performed by the state change unit 204c of FIG. First, the state changing unit 204c performs processing for specifying each value of the brightness coefficient z and the watermark coefficient x in the transformation matrix B from the received state change information (step S305).
[0106]
Next, the inverse matrix B of the transformation matrix B is used by using each value of the specified brightness coefficient z and watermark coefficient x. ^-1 Is generated (step S310).
[0107]
Then, a process for specifying the image with the digital watermark embedded, that is, the signed image Q is performed (step S315). This process is performed by designating an image file stored in the hard disk 210.
[0108]
Subsequently, variables i and j for image processing are set to an initial value 0 (steps S320 and 325). Then, for the variables i and j, it is determined whether or not the pixel qij of the signed image Q is a visible mark portion pixel (step S330).
[0109]
As a result of the determination, if the pixel qij is a non-marked pixel, the RGB component of the pixel qij is used as it is as the RGB component of the corresponding pixel qij 'of the new signed image Q' (step S340). ).
[0110]
On the other hand, when the pixel qij is a visible mark portion pixel, the RGB component of the pixel qij is converted into the inverse matrix B generated in step S310. ^-1 Thus, the RGB component of the corresponding pixel pij of the color image P, which is the original image, is obtained according to the equation (5) (step S335).
[0111]
[Equation 5]

[0112]
However, in actuality, when the RGB component of the pixel Pij in the color image P is converted by the conversion matrix B in the digital watermark embedding process shown in FIG. Since it is rounded to an integer, as described above, the RGB component of the pixel qij in the signed image Q is converted into the inverse matrix B of the transformation matrix B in step S335. ^-1 Even if converted by the above, the obtained value does not necessarily match the original RGB component of the pixel pij in the color image P that is the original image. That is, in the process of step S335, an approximate value of the original RGB component of the pixel pij in the color image P that is the original image is obtained.
[0113]
Next, the RGB component of the obtained pixel pij is set with the brightness coefficient z set to 1 and the watermark coefficient x set to the value specified in step S305 (that is, the same value as in the previous embedding process). The conversion matrix B is converted according to the above-described equation (3), and the obtained RGB component is set as the RGB component of the corresponding pixel qij ′ of the new signed image Q ′ (step S345).
[0114]
That is, in this case, it is expressed as in Expression (6).
[0115]
[Formula 6]

[0116]
Where B ₁ Represents a transformation matrix B with the brightness coefficient z set to 1, qij ¹ Is the transformation matrix B ₁ Represents the RGB component of the pixel obtained by conversion in (1). Specifically, qij ¹ The RGB components are expressed as shown in Equation (7).
[0117]
[Expression 7]

[0118]
Looking at the right side of Equation (7), in the first term, the RGB component (actual approximation value) of the pixel pij in the color image P that is the original image is maintained as it is, and in the second term, Only the color difference signal which is a watermark signal is newly added. As a result, the luminance component of each pixel is maintained as it is, and only the watermark information based on the watermark coefficient x is embedded as a watermark signal, so that the watermark information is embedded in an invisible state.
[0119]
That is, through the processing of steps S335 and S345, the pixel that was the visible mark portion is changed to the invisible state and becomes a pixel of the invisible mark portion.
[0120]
In step 345, as in the case of step S125, if the obtained RGB component is not an integer as a result of conversion by the conversion matrix B, the decimal part is rounded off to an integer.
[0121]
After performing the arithmetic processing as described above, the variables i and j are sequentially incremented (steps S350 and S360), and it is determined whether or not the processing has been completed for all the pixels of the watermark image W (steps S355 and S365). Until the completion, the above-described processing for changing the visible electronic watermark to the invisible state is repeated. When the process for all the pixels is completed, the digital watermark state changing process is terminated.
[0122]
Thereafter, when the user of the client computer 200 instructs display of an image, the CPU 202 of the client computer 200 reads out the image data in which the digital watermark is invisible from the hard disk 210 and causes the display device 212 to display the image data. At this time, the digital watermark is removed from the image displayed at this time. Therefore, the user can use the image without any inconvenience.
[0123]
Here, a specific example in which a digital watermark embedded in a visible state is made invisible will be briefly described.
[0124]
When the visible mark portion is made invisible in the procedure shown in FIG. 8 for the signed images shown in FIGS. 7A and 7B, the results are as shown in FIGS. 9A and 9B. A new signed image is obtained as shown. As is clear from FIG. 9, the mark portion displayed in the visible state becomes invisible, and at first glance, it appears as if the mark portion has been removed. Therefore, the user of the client computer 200 can use the purchased image without inconvenience.
[0125]
However, since the digital watermark is only invisible and not completely removed from the image, even if this image is illegally used, it can be easily detected and tracked.
[0126]
That is, when there is a suspicion that this image has been illegally used, the signed image shown in FIGS. 9A and 9B is used by using the color image which is the original image shown in FIGS. 5A and 5B. Then, the mark portion in the invisible state is extracted by a predetermined procedure. Specifically, the difference between the original color image and the signed image is taken for each RGB component, and the mark portion is extracted by binarizing the difference. As a result of the extraction, a logo mark image as shown in FIGS. 10A and 10B is obtained. As is clear from FIG. 10, the logo mark image embedded in the invisible state can be extracted to such an extent that the trace can be sufficiently confirmed.
[0127]
By the way, as described above, as a method of making the digital watermark embedded in the visible state invisible, there are a method of removing the digital watermark from the image data and a method of changing the digital watermark to the invisible state. In the above description, the case of using the latter method has been described. However, in the case of using the former method, the processing in the latter method is changed as follows.
[0128]
That is, in the latter method described above, in the state change process shown in FIG. 8, after obtaining the RGB component of the corresponding pixel pij of the color image P that is the original image (step S335), the digital watermark is changed to the invisible state. In order to do this, the RGB component of the pixel pij is converted to a value in which the brightness coefficient z is set to 1 and the watermark coefficient x is specified by the set conversion matrix B (step S345). Then, what is necessary is just to skip the process of this step S345.
[0129]
By doing so, the RGB component of the corresponding pixel pij of the color image P, which is the original image, becomes the RGB component of the corresponding pixel qij ′ of the image Q ′ as it is, so that the digital watermark is completely removed from the image Q ′. Will be removed.
[0130]
A-3. Effects of the first embodiment:
As described above, according to the present embodiment, image data is distributed from the server computer 100 to the client computer 200, and then the state change information is only distributed on condition that the image is purchased. The digital watermark embedded in the image can be made invisible, and the sample image can be easily converted into an image that can be used without any inconvenience. Therefore, it is not necessary for the server side (distributor side) to prepare sample data separately from the original image data. Further, it is not necessary for the client side (viewer side) to download the same image data twice.
[0131]
In addition, when the method of changing the visible watermark to the invisible state in order to make the watermark invisible, the watermark is not visible even after the watermark disappears from the image. It remains embedded. Therefore, even when an image is illegally used, it is possible to easily detect and track illegal use by detecting an embedded digital watermark.
[0132]
B. Second embodiment:
B-1. Information distribution system configuration:
FIG. 11 is an explanatory diagram showing the configuration of an information distribution system as a second embodiment of the present invention. In the first embodiment, the image data in which the digital watermark is embedded and the state change information for removing the digital watermark or changing the digital watermark to the invisible state are distributed from the same server computer 100. In the present embodiment, , To be distributed from different server computers. Therefore, in the information distribution system of this embodiment, a server computer 400 that distributes image data, a server computer 500 that distributes status change information, a client computer 200, and a charge management computer 350 are connected to each other. Network 300.
[0133]
Among these, the server computers 400 and 500 basically have the same configuration as the server computer 100 described in the first embodiment, but the computer programs stored in the main memories 404 and 504 are different. Yes.
[0134]
That is, the main memory 404 in the server computer 400 has functions of an electronic watermark embedding unit 404a that embeds an electronic watermark in image data in a visible state and an image data transmission unit 404b that transmits the image data embedded with the electronic watermark. While the computer program for realizing is stored, the main memory 504 in the server computer 500 is in a state change that transmits state change information for removing a visible digital watermark or changing it to an invisible state. A computer program for realizing the function of the information transmission unit 504a is stored. Therefore, since the other configuration is generally the same as that of the server computer 100, a detailed description is omitted.
[0135]
B-2. Information distribution process:
The information distribution process performed by the information distribution system of FIG. 11 is basically the same as the information distribution process in the first embodiment, except that the image data is from the server computer 400 and the state change information is from the server computer 500. It is different in that it is distributed separately to the client computer 200
Specifically, according to the flowchart shown in FIG. 2, first, in the server computer 400, the digital watermark embedding unit 404a performs a process of embedding the digital watermark in the visible state in the image data to be distributed (step). S12). Next, the CPU 402 of the server computer 400 notifies the client computer 200 of the presence of the image via the network 300 (step S14). In response to the notification, when the CPU 202 of the client computer 200 requests the server computer 400 to download an image (step S16), in the server computer 400, the image data transmission unit 404b makes the digital watermark visible. The embedded image data is transmitted to the client computer 200 via the network 300 (step S18).
[0136]
Thereafter, when the user of the client computer 200 decides to purchase an image and the accounting process (steps S22 and S24) is performed between the client computer 200 and the accounting management computer 350, the accounting management computer 350 Information on the completion of charging is sent to the server computer 500 via the network 300 (step S26). When the server computer 500 receives the information, the server computer 500 recognizes that the user of the client computer 200 has purchased the image, and the state change information transmission unit 504a changes the visible digital watermark to the invisible state. Is transmitted to the client computer 200 via the network 300 (step S28). At this time, the state change information transmission unit 504a encrypts the state change information, which is a secret key, and transmits it to the client computer 200.
[0137]
The subsequent processing is the same as in the case of the first embodiment, and a description thereof will be omitted. In the present embodiment, since the server computer 400 embeds the digital watermark in the image data, each value of the brightness coefficient z and the watermark coefficient x in the transformation matrix B that is the source of the state change information. Information initially exists in server computer 400. Therefore, in order to distribute the state change information from the other server computer 500 to the client computer 200, the information on the basis of the state change information and the information about the corresponding image are transmitted from the server computer 400 to the server computer 500. Need to be transmitted. As a specific transmission method, the information is encrypted and then transmitted via the network 300, or the information is directly transmitted via a dedicated line. The server computer 500 manages the state change information and the information related to the image received as described above in association with each other.
[0138]
B-3. Effects of the second embodiment:
As described above, also in the present embodiment, image data is distributed from the server computer 400 to the client computer 200, and thereafter, state change information is distributed from the server computer 500 to the client computer 200 on condition of image purchase. As a result, the digital watermark embedded in the image can be made invisible in the client computer 200, and the sample image can be easily converted into an image that can be used without any inconvenience. Accordingly, there is no need to prepare sample data separately from the original image data on the server side (distributor side), and the client side (viewer side) takes time to download the same image twice. There is an effect that it is not necessary.
[0139]
In this embodiment, the server computer 400 handles image data exclusively, and is responsible only for embedding digital watermarks and transmitting image data. The server computer 500 handles only state change information and changes state. Since it is only in charge of information management and transmission, the load on each server computer can be reduced. Since the load on the server computer 400 that handles image data is heavier than that on the server computer 500 that handles only state change information, a large number of server computers 400 that handle image data are prepared on the network 300 and obtained from them. The state change information may be managed by one server computer 500.
[0140]
C. Third embodiment:
C-1. Information distribution system configuration:
FIG. 12 is an explanatory diagram showing the configuration of an information distribution system as a third embodiment of the present invention. In the first and second embodiments, the image data in which the digital watermark is embedded is distributed from the server computer. However, in this embodiment, such image data is recorded on a CD-ROM, and the CD is recorded. The ROM is distributed to the users of the client computer 200. Therefore, in the information distribution system of this embodiment, a server computer that distributes image data is not necessary, and it is possible to read the image data recorded on the CD-ROM and the server computer 500 that distributes the state change information. A client computer 600 capable of receiving the distributed state change information, a charge management computer 350, and a network 300 to which these computers are connected.
[0141]
Among these, the client computer 600 basically has the same configuration as the client computer 200 described in the first embodiment, but the computer program stored in the main memory 604 is different. Specifically, in the main memory 604 of the client computer 600, a state change information receiving unit 604a that receives the state change information transmitted from the server computer 500, and embedded in the image data in a visible state based on the state change information. A computer program for realizing each function of the state changing unit 604b that removes the digital watermark that has been removed or changes the state to an invisible state is stored. Accordingly, since the other configuration is generally the same as that of the client computer 200, detailed description thereof is omitted.
[0142]
C-2. Information distribution process:
In this embodiment, as a premise, as described above, the image data is recorded on a CD-ROM and distributed to the user of the client computer 600. A digital watermark is embedded in the image data in advance in a visible state. As a method of distributing the CD-ROM, the CD-ROM may be distributed attached to a book or the like, or may be distributed directly in a store or on the street. Or you may make it deliver only to a specific user by mail or parcel delivery service.
[0143]
Thus, when the user of the client computer 600 obtains a CD-ROM on which image data is recorded, the user mounts the CD-ROM in the CD-ROM drive 616 in the same manner as the CD-ROM 618, Instructs reading and display of recorded image data. Then, the CPU 602 of the client computer 600 reads the image data from the CD-ROM by the CD-ROM drive 616 and displays it on the display device 612.
[0144]
As described above, since the digital watermark is embedded in the visible state in the image displayed at this time, the digital watermark interferes with the image, and it is inconvenient to use this image as it is. ing. That is, the image recorded on the CD-ROM is provided to the user as a sample.
[0145]
FIG. 13 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG. This process is realized as a process of the state change information transmission unit 504a, the state change information reception unit 604a, the state change unit 604b, and the like.
[0146]
When the user likes the displayed image and decides to purchase this image in order to use the image without any inconvenience and instructs the client computer 600 to do so, the information distribution process shown in FIG. 13 is started. First, the client computer 600 sends an image purchase instruction to the charge management computer 350 via the network 300 (step S42). The charging management computer 350 accesses the charging management information, identifies the user of the client computer 200, and charges the user for image purchase (step S44). At this time, if the user is a new user, the charging management computer 350 performs the above-described charging processing after the user registration procedure is completed.
[0147]
Next, billing management computer 350 sends billing completion information for the user to server computer 500 via network 300 (step S46). When the server computer 500 receives the information, it recognizes that the user of the client computer 200 has purchased the distributed image.
[0148]
When the server computer 500 recognizes the purchase of the image, the state change information transmission unit 504a transmits state change information for removing the digital watermark in the visible state or changing it to the invisible state to the client computer 600 via the network 300. (Step S48). At this time, the state change information transmission unit 104c encrypts the state change information, which is a secret key, and transmits it to the client computer 600.
[0149]
In the client computer 600, the state change information receiving unit 604a receives and decodes the transmitted state change information to obtain the original state change information (step S50). Thereafter, the state changing unit 604b reads out the image data to be embedded with the digital watermark from the CD-ROM by the CD-ROM drive 616, and based on the acquired state change information, the digital watermark is changed from the visible state. Processing to make the image invisible is performed (step S52), and the image data obtained thereby is stored in the hard disk 610.
[0150]
By following the above procedure, even if the image data is recorded on a distributed CD-ROM, the embedded digital watermark is changed from the visible state to the invisible state so that the image can be used without any inconvenience. It becomes possible to do.
[0151]
In this embodiment, a digital watermark is embedded in the image data in advance in a visible state, and the image data is recorded on a CD-ROM and distributed to the user. Therefore, the embedding of the digital watermark into the image data is basically performed at a place different from the server computer 500. Therefore, it is necessary to acquire information on each value of the brightness coefficient z and the watermark coefficient x in the transformation matrix B, which is the basis of the state change information, from the place where the embedding process is performed and input it to the server computer 500. is there. In inputting, not only the state change information but also information related to the image is input. The server computer 500 manages the state change information and the image-related information input in this manner in association with each other.
[0152]
C-3. Effects of the third embodiment:
As described above, in this embodiment, when the client computer 600 receives the distribution of the CD-ROM in which the image data is recorded, the state change from the server computer 500 to the client computer 200 is made on condition that the image is purchased. By only receiving information distribution, the digital watermark embedded in the image can be made invisible, and the sample image can be easily converted into an image that can be used without any inconvenience. Therefore, it is not necessary to prepare sample data separately from the original image data on the server side (distributor side), and on the client side (viewer side), in order to obtain the original image, There is no need to download data. Furthermore, when a method of changing a visible electronic watermark to an invisible state is employed, the electronic watermark remains embedded in an invisible state even after the electronic watermark is no longer visible from the image. Therefore, even when the image is illegally used, it is possible to easily detect and track the illegal use by detecting the digital watermark.
[0153]
D. Fourth embodiment:
D-1. Information distribution system configuration:
FIG. 14 is an explanatory diagram showing the configuration of an information distribution system as a fourth embodiment of the present invention. In the first embodiment, image data is distributed from the server computer 100, but in this embodiment, audio data is distributed from the server computer. Therefore, the information distribution system of this embodiment includes a server computer 700 that distributes various types of information such as audio data, a client computer 800 that can receive and reproduce the audio data, and a charge management computer 350. And a network 300 to which these computers are connected.
[0154]
Among them, the server computer 700 basically has the same configuration as the server computer 100 and the client computer 800 described in the first embodiment, but the main memories 704 and 804 are the same. The computer programs stored in are different.
[0155]
That is, in the main memory 704 in the server computer 700, a digital watermark embedding unit 704a that embeds a digital watermark in the target audio data in an audible state, an audio data transmission unit 704b that transmits the audio data in which the digital watermark is embedded, A computer program for realizing each function of the state change information transmission unit 704c that transmits state change information for removing the digital watermark in the audible state or making it inaudible is stored. The functions of the digital watermark embedding unit 704a, the audio data transmission unit 704b, and the state change information transmission unit 704c will be described in detail later.
[0156]
On the other hand, in the main memory 804 in the client computer 800, an audio data receiving unit 804a that receives audio data transmitted from the server computer, a state change information receiving unit 804b that receives state change information transmitted from the server computer, A computer program for realizing each function of the state changing unit 804c for removing the digital watermark embedded in the audio data in an audible state based on the state change information or changing it to an inaudible state is stored. The functions of the audio data receiving unit 804a, the state change information receiving unit 804b, and the state changing unit 804c will be described in detail later.
[0157]
Thus, in the present embodiment, since the target data is audio data, the digital watermark embedded in the data is information that can be perceived by hearing in a perceptible state. For example, noise and the like are included in addition to sounds, voices of people, music, and the like. In this specification, the electronic watermark is in an audible state means that the electronic watermark is in an audible state, and the inaudible state means a state in which the auditory recognition is difficult.
[0158]
D-2. Information distribution process:
FIG. 15 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG. This process is realized as a process such as a digital watermark embedding unit 704a, an audio data transmission unit 704b, a state change information transmission unit 704c, an audio data reception unit 804a, a state change information reception unit 804b, and a state change unit 804c.
[0159]
In the first embodiment described above, the transformation matrix B is used to embed a digital watermark in image data. However, in this embodiment, a discrete cosine transform (in order to embed a digital watermark in audio data) is used. DCT conversion) is used. Here, the DCT conversion will be briefly described.
[0160]
DCT transform is a method of orthogonal transform that transforms a signal into frequency components. The discrete cosine transform (DCT transform) and inverse transform (IDCT transform) are expressed by the following equations (8) to (10).
[0161]
[Equation 8]

[0162]
[Equation 9]

[0163]
[Expression 10]

[0164]
Here, g (n) is input data, G (m) is a conversion coefficient, and the size of the input data block to be converted is N.
[0165]
The DCT transform is generally used for two-dimensional images, such as MPEG, which is an international standard for moving image compression, and JPEG, which is an international standard for still image compression. It has the property of converting to a conversion coefficient with a certain property with little dependency. Since the DCT transform has such properties, when the watermark information is embedded in the transform coefficient, the watermark information is diffused in the entire image in an image obtained by inversely transforming the watermark information.
[0166]
Therefore, even if this DCT transform is applied to a one-dimensional voice, similarly, if watermark information is embedded in the transform coefficient and inversely transformed to obtain a voice, the watermark information is diffused throughout the voice. Will do.
[0167]
The transform coefficient for a block having a size of N is one DC component G (0) and N ² -1 AC component. In addition, the coefficient value of the DC component is maximized, and the coefficient value tends to be larger as the AC component is closer thereto. That is, most of the energy for forming an image is concentrated at a low frequency. At this time, the distribution of transform coefficient values for each block can be approximated by a Laplace distribution having a DC component at its peak. In order to embed watermark information in such a conversion coefficient in a natural manner, it is only necessary to embed the watermark information so that the Laplace distribution is not greatly changed.
[0168]
By the way, if the watermark information is DCT-transformed, most of the energy is concentrated at a low frequency and should have a distribution that can be approximated by a Laplace distribution. Therefore, if the conversion coefficient of the original speech and the conversion coefficient of the watermark information are multiplied by a predetermined ratio and added, the watermark information can be embedded in the conversion coefficient in a natural manner. Further, when this transform coefficient is inversely transformed into the spatial domain, the watermark information is dispersed throughout the speech. Therefore, it is difficult for a third party to illegally remove the digital watermark from the voice in which the digital watermark is embedded in both the frequency domain and the spatial domain.
[0169]
Now, returning to FIG. 15, when the information distribution process shown in FIG. 15 is started, first, in the server computer 700, the digital watermark embedding unit 704a applies a digital watermark to the audio data to be distributed. A process of embedding in an audible state is performed (step S62).
[0170]
Specifically, the digital watermark embedding unit 704a embeds a digital watermark according to the procedure of the embedding process shown in FIG. FIG. 16 is an explanatory diagram showing the procedure of the embedding process performed by the digital watermark embedding unit 104a of FIG. First, the digital watermark embedding unit 104a performs processing for specifying a voice (original voice) S to be embedded and a watermark voice T that is a digital watermark to be embedded. These processes are performed by designating an audio file stored in the hard disk 710.
[0171]
For the following discussion, U is a sound after embedding a watermark sound T as a digital watermark in the original sound S (hereinafter referred to as a signed sound).
[0172]
After preparing both voices S and T, next, the original voice S is DCT transformed to obtain a first DCT conversion coefficient Xs. Similarly, the watermark audio T is DCT-transformed to obtain a second DCT transform coefficient Xt.
[0173]
Next, according to the equation (11), corresponding frequency components of the first and second DCT transform coefficients are combined to generate a third DCT transform coefficient Xu.
[0174]
## EQU11 ##

[0175]
That is, for each corresponding frequency component, the first DCT coefficient Xs is multiplied by the weight coefficient a / n, and the second DCT coefficient Xt is multiplied by the weight coefficient b / m and added (or subtracted). 3 DCT transform coefficients Xu are generated. The two weighting factors a / n and b / m are arbitrary values other than 0. The two weight coefficients a / n and b / m are preferably rational numbers (that is, a, b, n and m are integers), and the sum is preferably set to 1. These two weighting factors a / n and b / m can be considered to represent relative ratios when the DCT coefficients of the original speech S and the watermark speech T are synthesized.
[0176]
Since the third DCT coefficient Xu includes watermark information, it is also referred to as “watermarked DCT coefficient”. When this watermarked DCT coefficient Xu is inversely transformed (IDCT), a signed speech U is obtained. That is, the signed voice U is a voice in which a watermark voice T, which is a digital watermark, is embedded in the original voice S in an audible state. Therefore, when the data of the signed voice U is reproduced, the watermark voice T and the original voice S are heard overlapping each other.
[0177]
As described above, when the signed voice, that is, the voice in which the digital watermark is audibly embedded, is obtained, the CPU 702 of the server computer 700 then transmits the presence of the voice to the network by using a Web page or an e-mail. The client computer 800 is informed via 300 (step S64). In response to the notification, if the user of the client computer 800 desires to download the sound, the CPU 802 of the client computer 800 requests the server computer 700 to download the sound in accordance with the instruction of the user (step S66).
[0178]
In the server computer 700, when the request is received, the audio data transmission unit 704b reads the audio data embedded with the digital watermark in an audible state from the hard disk 710, and transmits it to the client computer 800 via the network 300 (step S68). ). In the client computer 800, the audio data receiving unit 804a receives the audio data transmitted from the server computer 700 and stores it in the hard disk 810 (step S70).
[0179]
Then, when the user of the client computer 800 instructs the reproduction of the sound, the CPU 802 of the client computer 800 reads out the sound data from the hard disk 810 and causes the sound output device 814 to output it. At this time, since the digital watermark is audibly embedded in the reproduced audio, the user can hear the watermark audio in addition to the original audio as described above. Therefore, in order to listen to the original voice, the watermark voice becomes an obstacle and is inconvenient to use this voice. That is, in this state, this sound is provided to the user as a sample.
[0180]
Therefore, if the user likes this sound, it is necessary to purchase this sound in order to make the digital watermark embedded in an audible state inaudible.
[0181]
Therefore, when the user decides to purchase this voice and instructs the client computer 800 to do so, the client computer 800 sends a voice purchase instruction to the charging management computer 350 via the network 300 (step S72). When the management computer 350 receives the instruction, the management computer 350 identifies the user of the client computer 800 and charges the user for voice purchase (step S74).
[0182]
Then, billing management computer 350 sends billing completion information for the user to server computer 700 via network 300 (step S76). When the server computer 700 receives the information, it recognizes that the user of the client computer 800 has purchased the delivered voice.
[0183]
By the way, when an electronic watermark embedded in an audible state is made inaudible, a method of removing the digital watermark from audio data and a state in which the digital watermark remains in the audio data and is difficult to be recognized by hearing, that is, inaudible. It is possible to change the state. When the former method is used, the digital watermark no longer remains in the audio data, so that it is difficult to find and track the audio when it is illegally used. Therefore, the latter method is desirable from the viewpoint of prevention of unauthorized use and copyright protection.
[0184]
Therefore, in this embodiment, the latter method for changing the digital watermark from the visible state to the invisible state will be described first. In the server computer 700, when the voice purchase is recognized, the state change information transmitting unit 704c transmits state change information for making the audible digital watermark inaudible to the client computer 800 via the network 300 ( Step S78).
[0185]
Specifically, the state change information is each value of the second DCT coefficient Xt expressed by Equation (11) and the coefficient b / m−Δ approximated to the weight coefficient b / m. These values are secret keys that are known only to those who have embedded the digital watermark. By using these values, it is possible to change the audible digital watermark to the inaudible state. Accordingly, the state change information transmission unit 704c encrypts and transmits this state change information to the client computer 800 so that the state change information is not known to a third party.
[0186]
The state change information may be transmitted by using an electronic mail from the server computer 700 or by downloading from the client computer 800.
[0187]
In the client computer 800, the state change information receiving unit 804b receives the transmitted state change information, and decrypts the encrypted state change information (step S80). Then, the state changing unit 804c reads the audio data in which the digital watermark is embedded from the hard disk 810, and performs processing to change the digital watermark from the audible state to the inaudible state based on the decoded state change information ( Step S82).
[0188]
Specifically, the state changing unit 804c changes the state of the digital watermark according to the procedure of the state changing process shown in FIG. FIG. 17 is an explanatory diagram showing the procedure of the state change process performed by the state change unit 804c of FIG. First, the state changing unit 804c specifies each value of the second DCT coefficient Xt and the coefficient b / m−Δ approximated to the weighting coefficient b / m from the received state change information, and an electronic watermark is embedded. Voice, that is, signed voice U is specified. The signed voice U is specified by designating a voice file stored in the hard disk 810.
[0189]
Next, the state changing unit 804c obtains a third DCT transformation coefficient Xu by performing DCT transformation on the identified signed voice U. Then, with respect to the third DCT transform coefficient Xu, using the values of the second DCT coefficient Xt and the coefficient b / m−Δ specified earlier, the calculation shown in Expression (12) is performed, A DCT coefficient Xu ′ of 4 is obtained.
[0190]
[Expression 12]

[0191]
By this calculation, the watermark information ± (b / m) Xt included in the third DCT transform coefficient Xu is removed from the fourth DCT coefficient Xu ′, but a very small amount of ± ΔXt is newly added. Watermark information is included. Therefore, when the fourth DCT coefficient Xu ′ is inversely transformed (IDCT) to generate the signed voice U ′, only a small amount of the watermark voice T is embedded in the signed voice U ′. Even when the data of the signed voice U ′ is reproduced, only the original voice S can be heard, and it is difficult to recognize the watermark voice T by hearing. That is, the signed voice U ′ is a voice in which the watermark voice T, which is a digital watermark, is embedded in the original voice S in an inaudible state.
[0192]
In this way, when the state change process for changing the audible digital watermark to the inaudible state is completed, when the user of the client computer 800 instructs to reproduce audio, the CPU 802 of the client computer 800 cannot perform the digital watermark. The audio data in the listening state is read from the hard disk 810 and output to the audio output device 814. When the voice reproduced at this time is heard, it sounds as if the digital watermark has been removed. Therefore, the user can use this voice without any inconvenience.
[0193]
However, since the digital watermark is only inaudible and not completely removed from the voice, even if this voice is illegally used, it can be easily detected and tracked.
[0194]
That is, when there is a suspicion that this sound has been illegally used, the digital watermark embedded in an inaudible state is obtained by taking the difference between the original sound S and the signed sound U ′ using the original sound S. That is, the watermark sound T can be extracted to such an extent that the trace can be sufficiently confirmed.
[0195]
By the way, as described above, as a method of making the digital watermark embedded in an audible state inaudible, there are a method of removing the digital watermark from audio data and a method of changing the digital watermark to an inaudible state. In the above description, the case of using the latter method has been described. However, in the case of using the former method, the second DCT coefficient Xt and the weighting coefficient b / m shown in Expression (11) are used as state change information. In addition, the following processing is performed as the state change processing.
[0196]
Specifically, the state changing unit 804c changes the state of the digital watermark according to the procedure of the state changing process shown in FIG. FIG. 18 is an explanatory diagram showing another state change process performed by the state change unit 804c shown in FIG. First, the state changing unit 804c specifies each value of the second DCT coefficient Xt and the weighting factor b / m from the received state change information, and an electronic watermark is embedded from an audio file stored in the hard disk 810. Voice, that is, signed voice U is specified.
[0197]
Next, the state changing unit 804c obtains a third DCT transform coefficient Xu by performing DCT transform on the identified signed voice U, and further performs a second step on the third DCT transform coefficient Xu. Using the DCT coefficient Xt and the coefficient b / m, the calculation shown in Expression (13) is performed to obtain a fifth DCT coefficient Xs ′.
[0198]
[Formula 13]

[0199]
By this calculation, the watermark information ± (b / m) Xt included in the third DCT transform coefficient Xu is completely removed from the fifth DCT coefficient Xu ′. Therefore, when the fifth DCT coefficient Xs ′ is inversely transformed (IDCT) to generate the voice S ′, the voice S ′ is not embedded with the watermark voice T at all. Is played, only the original voice S can be heard. This sound S ′ is a sound in which the digital watermark embedded in the audible state is completely removed.
[0200]
D-3. Effects of the fourth embodiment:
As described above, according to the present embodiment, the client computer 800 simply distributes the audio data from the server computer 700 to the client computer 800 and then distributes the state change information on the condition of purchasing the audio. The digital watermark embedded in the audio in an audible state can be made inaudible, and the audio that has been a sample can be easily converted into audio that can be used without any inconvenience. Therefore, on the server side (distributor side), it is not necessary to prepare sample data separately from the original audio data. Further, it is not necessary for the client side (viewer side) to download the same audio data twice.
[0201]
In addition, in order to make the digital watermark inaudible, if a method of changing the audible digital watermark to the inaudible state is adopted, the digital watermark is inaudible even after the digital watermark is no longer heard from the sound. Since it remains embedded in the state, it is possible to easily detect and track unauthorized use by detecting the embedded digital watermark even if the sound is illegally used thereafter.
[0202]
E. Variations:
The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.
[0203]
E-1. Modification 1:
In the fourth embodiment described above, audio data is targeted, and from the server computer 700, audio data in which a digital watermark is embedded, state change information for removing the digital watermark or changing it to an inaudible state, Was to deliver. However, when audio data is targeted, the present invention is not limited to this embodiment, and similarly to the second embodiment when image data is targeted, audio data embedded with a digital watermark and digital watermark The state change information for removing or changing to the invisible state may be distributed from a different server computer. Similarly to the third embodiment, the audio data in which the digital watermark is embedded is CD- The data may be recorded on a ROM and the CD-ROM distributed to users of client computers.
[0204]
E-2. Modification 2:
In the first embodiment described above, the case where a logo mark image as shown in FIG. 6 is used as a watermark image serving as a digital watermark has been described as an example. However, the present invention is not limited to this.
[0205]
For example, a new effect can be expected if an image or sound carrying advertisements / promotions regarding products, services, companies, stores, etc. is used as a watermark image or watermark sound as a digital watermark. In other words, a provider of image data or audio data takes a sponsor and embeds the image or audio on which the advertisement or promotion relating to the sponsor is perceived in the image data or audio data and distributes it to the viewer. Or, by distributing, the image data and sound data itself can be a new advertisement / promotion medium. In addition, when the sponsor pays the advertisement / promotion cost to the provider, it is possible to distribute the state change information for removing the digital watermark or making it difficult to perceive to the viewer for free.
[0206]
E-3. Modification 3:
In the first embodiment described above, as shown in FIGS. 5 and 6, a still image is used as an image to be embedded with a digital watermark and as a watermark image to be a digital watermark. However, in the present invention, both the target image and the watermark image are not limited to still images, and moving images may be used.
[0207]
In the above description, the embodiment using image data as target data (ie, the first to third embodiments) and the embodiment using audio data (ie, the fourth embodiment) are divided. As described above, the present invention is naturally applicable to a case where image data such as a movie and audio data are used together as target data.
[0208]
E-4. Modification 4:
In the first embodiment described above, the digital watermark is embedded in the image data using the transformation matrix B shown in Equation (2). However, the method of embedding the digital watermark in the image data is limited to this. However, various methods can be used.
[0209]
For example, as described in Japanese Patent Application No. 11-170816 already filed by the applicant of the present application, RGB color system image data is converted into YCbCr colorimetry using a conversion matrix A shown in Expression (14). When converting into system image data, a watermark image is embedded using a transformation matrix X shown in Equation (15), and an inverse transformation matrix A shown in Equation (14) is used. ^-1 The signed image may be obtained by performing inverse conversion using.
[0210]
[Expression 14]

[0211]
[Expression 15]

[0212]
However, x is the watermark coefficient mentioned above.
[0213]
Similarly, as described in Japanese Patent Application No. 11-296648, which has already been filed by the applicant of the present application, image data is converted into four pixels using a transformation matrix C shown in Expression (16). A digital watermark may be embedded in the image data by conversion.
[0214]
[Expression 16]

[0215]
Here, d is a brightness coefficient that affects the luminance component of each pixel, and b1, b2, and b3 are watermark coefficients for generating watermark information.
[0216]
In the fourth embodiment, DCT conversion is used to embed a digital watermark in audio data. However, DCT conversion may also be used when a digital watermark is embedded in image data. In this case, as described in Japanese Patent Application No. 11-296204 already filed by the applicant of the present application, a two-dimensional DCT transformation is used as the DCT transformation.
[0217]
E-5. Modification 5:
In the embodiment described above, as a condition for the server computer to distribute the status change information to the client computer, the billing management computer 350 performs billing processing, and the server computer receives billing completion information from the billing management computer 350. However, the present invention is not limited to this, and various conditions can be used. .
[0218]
For example, it may be a condition that a user ID and a predetermined password are transmitted from the client computer to the server computer, or the user of the client computer makes a purchase by telephone or fax to the server computer administrator. It may be a condition that a will is shown.
[0219]
E-6. Modification 6:
In the third embodiment described above, image data embedded with a digital watermark is recorded on a CD-ROM and distributed. However, the present invention is not limited to this, and the digital watermark is embedded. Image data and audio data may be recorded and distributed on a flexible disk, hard disk, DVD-ROM, magneto-optical disk, IC card, ROM cartridge, magnetic tape cassette, or the like.
[0220]
E-7. Modification 7:
In the fourth embodiment described above, when generating the third DCT transform coefficient Xu, the corresponding frequency components of the first and second DCT transform coefficients are synthesized according to the equation (11). However, it is not always necessary to combine the frequencies corresponding to each other, and the frequencies may be combined with different frequencies. In that case, it is necessary to include information indicating which frequencies are combined in the status change information to be distributed.
In the fourth embodiment described above, the DCT transformation is used. However, other types of orthogonal transformation such as FFT and wavelet transformation can be employed.
[0221]
E-8. Modification 8:
In the first embodiment described above, in the server computer 100, in the second embodiment, in the server computer 400, in the fourth embodiment, the server computer 700 embeds an electronic watermark in image data or audio data, respectively. However, the present invention is not limited to this, and embedding of a digital watermark may be performed at a place other than the server computer. In other words, the server computer only needs to be able to store image data and audio data embedded with such digital watermarks and distribute them to client computers.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram showing a configuration of an information distribution system as a first embodiment of the present invention.
FIG. 2 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG. 1;
3 is a flowchart showing a procedure of an embedding process performed by the digital watermark embedding unit 104a of FIG.
FIG. 4 is an explanatory diagram for explaining RGB components of a pixel Pij in a color image P to be embedded.
FIG. 5 is an explanatory diagram illustrating an example of a color image P to be embedded.
FIG. 6 is an explanatory diagram showing an example of a watermark image W to be embedded.
7 is an explanatory diagram showing an example of a signed image obtained by embedding the watermark image W of FIG. 6 in the color image P of FIG.
FIG. 8 is a flowchart showing a procedure of state change processing performed by the state change unit 204c of FIG.
9 is an explanatory diagram showing an example of a new signed image obtained by making a visible mark portion invisible in the signed image of FIG.
10 is an explanatory diagram showing an example of a logo mark image obtained by extracting a mark portion that is invisible from the signed image in FIG. 9;
FIG. 11 is an explanatory diagram showing a configuration of an information distribution system as a second embodiment of the present invention.
FIG. 12 is an explanatory diagram showing a configuration of an information distribution system as a third embodiment of the present invention.
13 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG.
FIG. 14 is an explanatory diagram showing a configuration of an information distribution system as a fourth example of the present invention.
15 is a flowchart showing a procedure of information distribution processing performed by the information distribution system of FIG. 14;
16 is an explanatory diagram showing a procedure of an embedding process performed by the digital watermark embedding unit 104a of FIG.
FIG. 17 is an explanatory diagram showing a procedure of state change processing performed by the state change unit 804c of FIG. 14;
FIG. 18 is an explanatory diagram showing another state change process performed by the state change unit 804c of FIG. 14;
[Explanation of symbols]
100: Server computer
102 ... CPU
104 ... main memory
104a ... Digital watermark embedding unit
104b ... Image data transmission unit
104c ... Status change information transmission unit
106 ... Keyboard
108 ... Mouse
110: Hard disk
112 ... Display device
114 ... Audio output device
116, 216 ... CD-ROM drive
116 ... CD-ROM drive
118 ... CD-ROM
120: Communication device
122 ... Bus
200: Client computer
202 ... CPU
204 ... main memory
204a ... Image data receiving unit
204b ... Status change information receiving unit
204c ... state change part
206 ... Keyboard
208 ... Mouse
210 ... Hard disk
212 ... Display device
214 ... Audio output device
216 ... CD-ROM drive
218 ... CD-ROM
220: Communication device
222 ... Bus
300 ... Network
350 ... Billing management computer
400: Server computer
402 ... CPU
404 ... main memory
404a ... Digital watermark embedding unit
404b ... Image data transmission unit
406 ... Keyboard
408 ... mouse
410: Hard disk
412 ... Display device
414 ... Audio output device
416 ... CD-ROM drive
418 ... CD-ROM
420: Communication device
422 ... Bus
500 ... Server computer
504 ... Main memory
504a: State change information transmitter
506 ... Keyboard
508 ... Mouse
510: Hard disk
512. Display device
514 ... Audio output device
516 ... CD-ROM drive
518 ... CD-ROM
520 ... Communication device
522 ... Bus
600: Client computer
602 ... CPU
604 ... Main memory
604a: State change information receiving unit
604b ... state change part
606 ... Keyboard
608 ... Mouse
610: Hard disk
612 ... Display device
614 ... Audio output device
616 ... CD-ROM drive
618 ... CD-ROM
620 ... Communication device
622 ... Bus
700: Server computer
702 ... CPU
704 ... Main memory
704a: Digital watermark embedding unit
704b ... Audio data transmission unit
704c ... Status change information transmission unit
706 ... Keyboard
708 ... mouse
710: hard disk
712 ... Display device
714 ... Audio output device
716 ... CD-ROM drive
718 ... CD-ROM
720 ... Communication device
722 ... Bus
800: Client computer
802 ... CPU
804 ... Main memory
804a: Audio data receiving unit
804b ... Status change information receiving unit
804c: State change unit
806 ... Keyboard
808 ... mouse
810: Hard disk
812 ... Display device
814 ... Audio output device
816 ... CD-ROM drive
818 ... CD-ROM
820 ... Communication device
822 ... Bus

Claims (26)

An information distribution system for distributing information from a server computer to a client computer,
The server computer
A digital watermark embedding unit that embeds a perceptible digital watermark in audio data;
A data transmitting unit that transmits the target data the watermark is voice data padded, to the client computer,
When a predetermined condition is satisfied, and the against the client computer, pre SL state change information transmitting unit that transmits the state change information to or perceive a difficult state to remove the watermark,
With
The digital watermark embedding unit includes:
DCT transforming the audio data to generate a first DCT coefficient Xs;
DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xs multiplied by the first weighting coefficient (a / n) and the second DCT coefficient Xt multiplied by the second weighting coefficient (b / m). The multiplied product is added to generate a third DCT coefficient Xu,
The third DCT coefficient Xu is subjected to inverse DCT transform to generate the target data in which the digital watermark is embedded,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, a, b, m and n are integers, and Δ is a value smaller than b / m.
The client computer is
A data receiving unit for receiving the target data transmitted from the server computer;
A state change information receiving unit for receiving the state change information transmitted from the server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
Equipped with a,
The state change unit
DCT transforming the target data to restore the third DCT coefficient Xu,
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). By subtracting from the DCT coefficient, a fourth DCT coefficient Xu ′ is generated,
Inverse DCT transform the fourth DCT coefficient Xu ′;
The digital watermark is removed, or voice data embedded with the digital watermark being difficult to perceive is generated.
Information distribution system. An information distribution system for distributing information from a first and second server computer to a client computer,
The first server computer is
A digital watermark embedding unit that embeds a perceptible digital watermark in audio data;
The target data is audio data that the digital watermark is padded, and a data transmission unit to be transmitted to the client computer,
The digital watermark embedding unit includes:
DCT transforming the audio data to generate a first DCT coefficient Xs;
DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xu is multiplied by a first weighting coefficient (a / n), and the second DCT coefficient Xt is multiplied by a second weighting coefficient (b / m). The multiplied product is added to generate a third DCT coefficient Xu,
The third DCT coefficient Xu is subjected to inverse DCT transform to generate the target data in which the digital watermark is embedded,
Here, a, b, m, and n are integers,
The second server computer is
When a predetermined condition is satisfied, against the client computer, a status change information to or perceive a difficult state to remove the watermark, the second weighting factor (b / m) And the third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m), and the second DCT coefficient Xt , A state change information transmitter for transmitting,
Here, Δ is a value smaller than b / m,
The client computer is
A data receiving unit for receiving the target data transmitted from the first server computer;
A state change information receiving unit for receiving the state change information transmitted from the second server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
Equipped with a,
The state change unit
DCT transforming the target data to restore the third DCT coefficient Xu,
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). By subtracting from the DCT coefficient, a fourth DCT coefficient Xu ′ is generated,
Inverse DCT transform the fourth DCT coefficient Xu ′;
Accordingly, the information distribution system generates the audio data in which the digital watermark is removed or the digital watermark is embedded in a state where it is difficult to perceive . An information distribution system for distributing information from a server computer to a client computer having target data that is audio data embedded in a state where a digital watermark can be perceived,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The server computer
When a predetermined condition is satisfied, against the client computer, a status change information to or perceive a difficult state to remove the watermark, the second weighting factor (b / m) And the third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m), and the second DCT coefficient Xt , A state change information transmitter for transmitting,
Here, Δ is a value smaller than b / m,
The client computer is
A state change information receiving unit for receiving the state change information transmitted from the server computer;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
Equipped with a,
The state change unit
DCT transforming the target data to restore the third DCT coefficient Xu,
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). By subtracting from the DCT coefficient Xu, a fourth DCT coefficient Xu ′ is generated,
Inverse DCT transform the fourth DCT coefficient Xu ′;
Accordingly, the information distribution system generates the audio data in which the digital watermark is removed or the digital watermark is embedded in a state where it is difficult to perceive . In the information delivery system according to claim 3,
The information distribution system according to claim 1, wherein the target data of the client computer is recorded on a recording medium readable by the client computer. In the information delivery system according to any one of claims 1 to 4,
The electronic watermark includes at least one of advertisement information and advertisement information. In the information delivery system according to any one of claims 1 to 5,
The information distribution system according to claim 1, wherein the predetermined condition is that the server computer recognizes that a voice of the target data has been purchased by a user of the client computer. In the information delivery system according to any one of claims 1 to 6,
The state change information transmission unit encrypts and transmits the state change information,
The information distribution system, wherein the state change information receiving unit receives and decrypts the encrypted state change information. In the information delivery system according to any one of claims 1 to 7,
The third DCT transform coefficient Xu is
Xu = (a / n) .Xs ± (b / m) .Xt
Is generated by the operation of
The fourth DCT transform coefficient Xu ′ is
Xu ′ = (a / n) · Xs ± Δ · Xt
Generated by the operation of
Information distribution system. In the information delivery system according to any one of claims 1 to 8 ,
The first weighting factor (a / n) and the second weighting factor (b / n) are both non-zero rational numbers, and the first weighting factor (a / n) and the second weighting factor (b / M) An information distribution system in which the sum of 1 is 1. A server computer for delivering information to a client computer,
A digital watermark embedding unit that embeds a perceptible digital watermark in audio data;
A data transmission unit that transmits target data, which is audio data embedded in a state where the digital watermark is perceptible, to the client computer;
A state change information transmitting unit that transmits state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied;
Equipped with a,
The digital watermark embedding unit includes:
DCT transforming the audio data to generate a first DCT coefficient Xs;
DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xs multiplied by the first weighting coefficient (a / n) and the second DCT coefficient Xt multiplied by the second weighting coefficient (b / m). The multiplied product is added to generate a third DCT coefficient Xu,
The third DCT coefficient Xu is subjected to inverse DCT transform to generate the target data in which the digital watermark is embedded,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, the a, b, m, n are integers, wherein Δ is Ru value smaller der than b / m a server computer. A client computer that receives information distributed from a server computer,
A data receiving unit that is transmitted from the first server computer, the electronic watermark to receive target data is audio data embedded in perceptible state,
Status change for removing the perceptible digital watermark transmitted from the first server computer or a second server computer different from the first server computer or making it difficult to perceive A state change information receiver for receiving information;
A state change unit that removes or changes the digital watermark embedded in the target data in a perceptible state to a state that is difficult to perceive based on the state change information;
Equipped with a,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, Δ is a value smaller than b / m,
The state change unit
DCT transforming the target data to restore the third DCT coefficient Xu,
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). By subtracting from the DCT coefficient Xu, a fourth DCT coefficient Xu ′ is generated,
Inverse DCT transform the fourth DCT coefficient Xu ′;
The digital watermark is removed, or the voice data embedded with the digital watermark being difficult to perceive is generated.
Client computer. A server computer for delivering information to a client computer,
A digital watermark embedding unit that embeds a perceptible digital watermark in audio data;
The target data the watermark is voice data embedded in perceptible state, a data transmission unit to be transmitted to the client computer,
The digital watermark embedding unit includes:
DCT transforming the audio data to generate a first DCT coefficient Xs;
DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xs multiplied by the first weighting coefficient (a / n) and the second DCT coefficient Xt multiplied by the second weighting coefficient (b / m). The multiplied product is added to generate a third DCT coefficient Xu,
The third DCT coefficient Xu is subjected to inverse DCT transform to generate the target data in which the digital watermark is embedded,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, the a, b, m, n are integers, delta is b / m value smaller der than Ru server computer. A server computer that distributes information to a client computer having target data that is audio data embedded in a state where a digital watermark can be perceived,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
When a predetermined condition is satisfied, against the client computer, a status change information to or perceive a difficult state to remove the watermark, the second weighting factor (b / m) And the third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m), and the second DCT coefficient Xt , A server computer comprising a state change information transmission unit for transmission. A client computer having target data, which is audio data embedded in a state where a digital watermark can be perceived, and receiving information distributed from a server computer,
A state change information receiving unit that receives state change information transmitted from the server computer for removing the digital watermark in a perceptible state or making it difficult to perceive;
Based on the state change information, a state change unit that changes the digital watermark embedded in the data to be recorded from a perceptible state to a difficult perceptible state;
Equipped with a,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
The state change unit
DCT transforming the target data to restore the third DCT coefficient Xu,
A coefficient generated by multiplying either the second weighting coefficient (b / m) or the third weighting coefficient (b / m-Δ) and the second DCT coefficient is defined as the third DCT. By subtracting from the coefficient Xu, a fourth DCT coefficient Xu ′ is generated,
Inverse DCT transform the fourth DCT coefficient Xu ′;
It allows the watermark is removed, or the client computer said electronic watermark that generates the audio data embedded in difficult conditions perception. An information distribution method for distributing information from a server computer to a client computer,
(I) a step of embedding a digital watermark for embedding a perceptible digital watermark in audio data;
(Ii) transmitting target data, which is audio data embedded in a state where a digital watermark can be perceived, from the first server computer to the client computer;
( Iii ) When a predetermined condition is satisfied, the digital watermark in a perceivable state is sent from the first server computer or a second server computer different from the first server computer to the client computer. Transmitting state change information for removal or rendering it difficult to perceive;
( Iv ) in the client computer, removing the digital watermark embedded in a perceptible state in the target data based on the state change information or changing the digital watermark to a state that is difficult to perceive;
With
The step (i)
DCT transforming the audio data to generate a first DCT coefficient Xs;
DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xs multiplied by the first weighting coefficient (a / n) and the second DCT coefficient Xt multiplied by the second weighting coefficient (b / m). Adding the multiplied product to generate a third DCT coefficient Xu;
Generating the target data embedded with the digital watermark by performing an inverse DCT transform on the third DCT coefficient Xu,
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, Δ is a value smaller than b / m,
The step (iv) includes
DCT transforming the target data to restore the third DCT coefficient Xu;
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). A step of generating a fourth DCT coefficient Xu ′ by subtracting from the DCT coefficient Xu;
Generating the audio data in which the digital watermark is removed or the digital watermark is embedded in a state in which the digital watermark is difficult to perceive by performing an inverse DCT transform on the fourth DCT coefficient Xu ′;
Including information delivery method. An information distribution method for distributing information from a server computer to a client computer having target data that is audio data embedded in a state where a digital watermark can be perceived,
( I ) When a predetermined condition is satisfied, state change information for removing the digital watermark in a perceptible state or making it difficult to perceive is transmitted from the server computer to the client computer. Process,
( Ii ) in the client computer, removing the digital watermark embedded in a perceptible state in the target data based on the state change information or changing the digital watermark to a state that is difficult to perceive;
Equipped with a,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, Δ is a value smaller than b / m,
The step (ii)
DCT transforming the target data to restore the third DCT coefficient Xu;
A coefficient generated by multiplying either the second weight coefficient (b / m) or the third weight coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weight coefficient (b / m). A step of generating a fourth DCT coefficient Xu ′ by subtracting from the DCT coefficient Xu;
Generating the audio data in which the digital watermark is removed or the digital watermark is embedded in a state in which the digital watermark is difficult to perceive by performing an inverse DCT transform on the fourth DCT coefficient Xu ′;
including,
Information distribution method. The information delivery method according to claim 16, wherein
The target data possessed by the client computer is recorded on a recording medium readable by the client computer. The information delivery method according to any one of claims 15 to 17,
The electronic watermark includes at least one of advertisement information and advertisement information. In the information delivery method according to any one of claims 15 to 18,
The information distribution method according to claim 1, wherein the predetermined condition is that the server computer recognizes that a user of the client computer has purchased an image or sound represented by the target data. In the information delivery method according to any one of claims 15 to 19,
The step ( i )
In the server computer, the step of encrypting and transmitting the state change information;
Receiving and decrypting the encrypted state change information at the client computer;
An information delivery method comprising: The information delivery method according to any one of claims 15 to 20,
The third DCT transform coefficient Xu is
Xu = (a / n) .Xs ± (b / m) .Xt
Is generated by the operation of
The fourth DCT transform coefficient Xu ′ is
Xu ′ = (a / n) · Xs ± Δ · Xt
Generated by the operation of
An information delivery method characterized by the above. The information delivery method according to any one of claims 15 to 21, wherein
The first weighting factor (a / m) and the second weighting factor (b / n) are both non-zero rational numbers, and the first weighting factor (a / m) and the second weighting factor (b / N) is 1, and the information distribution method is characterized in that A computer-readable recording medium recording a computer program for distributing information to a client computer,
( I ) a function of embedding a perceptible digital watermark in audio data;
(Ii) a function of transmitting target data, which is audio data embedded in a state where a digital watermark can be perceived, to the client computer;
( Iii ) a function of transmitting state change information for removing the digital watermark in a perceptible state or making it difficult to perceive to the client computer when a predetermined condition is satisfied;
Is recorded a computer program to make the computer realized ,
The function (i) is
A function of DCT transforming the audio data to generate a first DCT coefficient Xs;
A function of DCT transforming the digital watermark to generate a second DCT coefficient Xt;
For each corresponding frequency component, the first DCT coefficient Xs is multiplied by a first weighting coefficient (a / n), and the second DCT coefficient Xt is multiplied by a second weighting coefficient (b / m). A function of adding the multiplied product to generate a third DCT coefficient Xu;
A function of performing inverse DCT on the third DCT coefficient Xu to generate the target data in which the digital watermark is embedded;
Including
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Δ is a value smaller than b / m.
recoding media. A computer-readable recording medium storing a computer program for receiving information distributed from a server computer,
(I) a function of receiving target data, which is audio data embedded in a state where a digital watermark can be perceived, transmitted from the first server computer;
(Ii) to remove the perceptible digital watermark transmitted from the first server computer or a second server computer different from the first server computer, or to make it difficult to perceive A function for receiving state change information;
(Iii) a function of removing the digital watermark embedded in the perceptible state in the target data based on the state change information or changing the digital watermark to a state that is difficult to perceive;
Recorded thereon a computer program for realizing the to the computer,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, Δ is a value smaller than b / m,
The function (iii) is
A function of DCT transforming the target data to restore the third DCT coefficient;
A coefficient generated by multiplying either the second weighting coefficient (b / m) or the third weighting coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weighting coefficient (b / m). A function of generating a fourth DCT coefficient Xu ′ by subtracting from the DCT coefficient Xu;
A function of inverse DCT transforming the fourth DCT coefficient Xu ′;
A function of generating the audio data in which the digital watermark is removed or embedded with the digital watermark being difficult to perceive,
recoding media. A computer-readable recording medium recording a computer program for distributing information to a client computer having target data that is audio data embedded in a state where a digital watermark can be perceived,
To cause a computer to realize a function of transmitting state change information for removing the digital watermark in a perceptible state or making it difficult to perceive when a predetermined condition is satisfied Record computer programs ,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The function includes, as state change information, the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The ability to send one of the
Sending the second DCT coefficient Xt,
Here, Δ is a value smaller than b / m.
recoding media. A computer-readable recording medium recording a computer program for receiving information distributed from a server computer by a client computer having target data which is audio data embedded in a state where a digital watermark can be perceived,
( I ) a function of receiving state change information transmitted from the server computer for removing the digital watermark in a perceptible state or making it difficult to perceive;
( Ii ) a function of changing the electronic watermark embedded in the data to be recorded from a perceptible state to a difficult to perceive state based on the state change information;
Is recorded a computer program for causing the client computer to realize,
The target data is
A first DCT coefficient Xs generated by DCT conversion of the audio data and a first weighting coefficient (a / n), and a second DCT coefficient generated by DCT conversion of the digital watermark A third DCT coefficient Xu generated by adding Xt multiplied by the second weighting coefficient (b / m) for each corresponding frequency component and further generated by performing inverse DCT conversion. And
Here, a, b, m, and n are integers,
The state change information includes one of the second weighting factor (b / m) and a third weighting factor (b / m−Δ) approximated to the second weighting factor (b / m). The second DCT coefficient Xt,
Here, Δ is a value smaller than b / m,
The function (ii) is
A function of DCT transforming the target data to restore the third DCT coefficient Xu;
A coefficient generated by multiplying either the second weighting coefficient (b / m) or the third weighting coefficient (b / m−Δ) and the second DCT coefficient Xt is the third weighting coefficient (b / m). A function of generating a fourth DCT coefficient Xu ′ by subtracting from the DCT coefficient Xu;
A function of inverse DCT transforming the fourth DCT coefficient Xu ′;
A function of generating the audio data in which the digital watermark is removed or embedded with the digital watermark being difficult to perceive,
recoding media.

Images