JP2004344237A - Device and method for measuring visual load, program for measuring visual load and recording medium recording the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method, a device, a program and a recording medium for measuring a visual load felt by an observer against color blinking light. <P>SOLUTION: The device for measuring the visual load consists of a color characteristic input part 1, a blinking light preparing part 2, a blinking light presentation part 3, a blinking light measuring part 4, a visual load measuring part 5, and a visual load output part 6. The light characteristic input part 1 inputs color characteristics of two colors constituting the blinking light and the time-frequency of the blinking light. The blinking light is prepared by the color blinking light preparing part 2 based on the inputted color characteristics. The prepared blinking light is presented by the blinking light presenting part 3 by alternately presenting frames (visual screens) having the previously inputted two colors with the inputted time-frequency. The color characteristics of the blinking light presented by the blinking light presenting part 3 are presented by the color characteristic measuring part 4. The visual load is calculated by the visual load calculating part 5 based on the cone response quantity of the two colors constituting the blinking light using the color characteristic obtained. The visual load is also similarly calculated by directly transmitting the color characteristic value inputted by the color characteristic input part 1 to the visual load calculating part 5. The visual load calculated in this way is presented to a user by the visual load output part 6. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

ここで
【００１１】
【数２】

である。
【００１２】
式（１）のＬＭは式（２）で表わされるＬＭ反対色信号における二色間の差分、式（１）のＳは式（３）で表わされるＳ反対色信号における二色間の差分、式（１）のＬｕｍｉｎａｎｃｅは式（４）で表わされる二色間の輝度の差分である。詳細は後述する。α、β、γはＬＭ、Ｓ、Ｌｕｍｉｎａｎｃｅそれぞれの項の重み付け係数を示す。式（２）、（３）、（４）中のＬ_１、Ｍ_１、Ｓ_１は点滅光を構成する一方の色がもたらすＬ錐体、Ｍ錐体、Ｓ錐体の応答量、Ｌ_２、Ｍ_２、Ｓ_２はもう片方の色がもたらすＬ錐体、Ｍ錐体、Ｓ錐体の応答量である。また、ｆ（ｔ）は二色間の色差と輝度差によりもたらされる効果が、点滅光の時間周波数（１秒間に点滅する回数）によりどのように変化するか示す関数であり、式（５）で表わされる。ここで、ｔは点滅光の時間周波数（単位Ｈｚ）、ｔ_ｐｅａｋは点滅光の効果がピークとなる時間周波数、ｓｄは関数の広がりを示す標準偏差である。重み付け係数（α、β、γ）と時間周波数の効果（ｆ（ｔ））は後述する実験により求めた。その結果、重み付け係数は
α＝１
β＝０．２
γ＝１
となり、また時間周波数の関数ｆ（ｔ）における２つのパラメータはそれぞれ
ｔ_ｐｅａｋ＝８
ｓｄ＝４
となる。
【００１３】
人間の網膜には、錐体と呼ばれる３種類の光受容細胞（Ｌ、Ｍ、Ｓ錐体）が分布している。錐体の応答量は、視覚パターンを観察者が見たときに、その観察者の網膜に存在する錐体が活性化する量である。３錐体はそれぞれ異なる分光特性を持つため、光刺激が与えられると、それに対して独自の応答をする。この３錐体からの応答が私たちの色覚の基礎となる。
【００１４】
各錐体の応答量は、各錐体の分光感度特性（各錐体の各波長に対する感度関数）と、画面上に提示される色の分光放射輝度から求めることができる。各錐体の分光感度特性には大きな個人差がないため、一般には基本感度関数と呼ばれる平均値が用いられる（Ｖ． Ｃ． Ｓｍｉｔｈ， Ｊ． Ｐｏｋｏｒｎｙ， Ｖｉｓｉｏｎ Ｒｅｓｅａｒｃｈ， Ｖｏ．１５， ｐｐ．１６１−１７１ （１９７５））。錐体の応答量（Ｌ、Ｍ、Ｓ）は、次式（６）で表される。
【００１５】
【数３】

ここで、ｌ（λ）、ｍ（λ）、ｓ（λ）は各錐体の基本感度関数、Ｃ（λ）は画面に提示された色の分光放射輝度を表す。また、画面上に提示される色の分光放射輝度が計測できない場合でも、色のｘｙ色度座標値（ｘ，ｙ）と輝度値（Ｌｕｍ）を測定すれば（あるいは画面に色提示する前にあらかじめ定めておけば）、錐体応答量を計算することができる。まず、ｘｙ色度座標値（ｘ，ｙ）と輝度値（Ｌｕｍ）をＸＹＺ表色系における三刺激値Ｘ、Ｙ、Ｚに変換する。
【００１６】
【数４】

そして、三刺激値Ｘ、Ｙ、Ｚから錐体の応答量を次式（８）により求めることができる。
【００１７】
【数５】

ここで、ｋは正規化のための定数、また（ｘ_ｌ、ｙ_ｌ、ｚ_ｌ）（ｘ_ｍ、ｙ_ｍ、ｚ_ｍ）（ｘ_ｓ、ｙ_ｓ、ｚ_ｓ）はそれぞれＬ、Ｍ、Ｓ各錐体のｘｙ色度座標における混同線の収束点であり、Ｌ、Ｍ、Ｓ各錐体をそれぞれ一意に同定するｘｙ色度値を意味する。これらの値はＳｍｉｔｈ ＆ Ｐｏｋｏｒｎｙ（１９７５）に記されている。
【００１８】
また、画面に提示する色をあらかじめＲＧＢ値で規定する場合でも、錐体の応答量を算出することができる。まず、次式（９）により、三刺激値Ｘ、Ｙ、Ｚを求める。
【００１９】
【数６】

（ｘ_Ｒ、ｙ_Ｒ、ｚ_Ｒ）、（ｘ_Ｇ、ｙ_Ｇ、ｚ_Ｇ）、（ｘ_Ｂ、ｙ_Ｂ、ｚ_Ｂ）は使用する画像提示装置におけるＲ、Ｇ、Ｂ信号のｘｙ色度座標値である。これらの値は使用する装置により異なるため、あらかじめ測定しておくことが必要になる。（Ｌ_Ｒ、Ｌ_Ｇ、Ｌ_Ｂ）は、点滅光の各色をもたらすＲ、Ｇ、Ｂ各信号の輝度値である。式（９）から得られた三刺激値Ｘ、Ｙ、Ｚを式（８）に代入し、三錐体（Ｌ、Ｍ、Ｓ）の応答量を求めることができる。
【００２０】
以上のように、錐体応答量（Ｌ_１、Ｍ_１、Ｓ_１、Ｌ_２、Ｍ_２、Ｓ_２）は、色特性測定部により測定された点滅光を構成する各色の諸特性（点滅光各色のｘｙ色度座標値、点滅光各色の輝度値、点滅光各色の分光感度特性）と、既知である３錐体の分光感度特性（Ｓｍｉｔｈ ＆ Ｐｏｋｏｒｎｙ， １９７５）から計算できる。あるいは、色特性入力部において、あらかじめ入力された色特性値（画像提示装置のＲＧＢ信号のｘｙ色度座標値、点滅光各色のＲＧＢそれぞれの輝度値）から錐体応答量を計算することもできる。
【００２１】
点滅光を構成する二色それぞれが３錐体にもたらす応答量（Ｌ_１、Ｍ_１、Ｓ_１、Ｌ_２、Ｍ_２、Ｓ_２）と点滅光の時間周波数（ｔ）がわかれば、式（１）から視覚負荷度を算出することができる。式（１）は、点滅光の二色間における色差と輝度差を示す項を持つ。式（１）のＬＭ項は、Ｐａｒｖｏ細胞経路という神経細胞群が運ぶＬＭ反対色と呼ばれる色信号（Ｌ錐体とＭ錐体の応答の差分）の差分である（Ｄｅ Ｖａｌｏｉｓ ＆ Ｄｅ Ｖａｌｏｉｓ， ２０００， Ｓｅｅｉｎｇ， Ａｃａｄｅｍｉｃ Ｐｒｅｓｓ）。点滅光を構成する一方の色のＬＭ反対色成分は（Ｌ_１−Ｍ_１）／（Ｌ_１＋Ｍ_１）、もう片方の色のＬＭ反対色成分は（Ｌ_２−Ｍ_２）／（Ｌ_２＋Ｍ_２）である。式（１）のＬＭ項では、このＬＭ反対色成分の差分の絶対値を算出する。式（１）のＳ項は、Ｋｏｎｉｏ細胞経路という神経細胞群が運ぶＳ反対色と呼ばれる色信号（Ｌ錐体とＭ錐体の応答の和とＳ錐体の応答との差分）の差分である。点滅光を構成する一方の色のＳ反対色成分はＳ_１−（Ｌ_１＋Ｍ_１）、もう片方の色のＳ反対色成分はＳ_２−（Ｌ_２＋Ｍ_２）である。式（１）のＳ項では、このＳ反対色成分の差分の絶対値を算出する。式（１）のＬｕｍｉｎａｎｃｅ項はＭａｇｎｏ細胞経路という神経細胞群が運ぶ輝度信号（Ｌ錐体とＭ錐体の応答の和）の差分である。点滅光を構成する一方の色の輝度成分はＬ_１＋Ｍ_１、もう片方の色の輝度成分はＬ_２＋Ｍ_２である。式（１）のＬｕｍｉｎａｎｃｅ項では、この輝度成分の差分の絶対値を算出する。
【００２２】
視覚負荷度に対して、これらの３つの成分（ＬＭ、Ｓ、Ｌｕｍｉｎａｎｃｅ）がどのような割合で影響するか（α、β、γ）、また点滅光の頻度（時間周波数ｔ）がどのように影響するか、を以下に示す心理実験に基づいて導出した。まず、ある輝度コントラストを持つパターン（正弦波格子縞パターン）をコンピュータで作成する。続いて、点滅光を構成する背景色を二色選び、観察者に対してパターンと点滅光を同時に提示する（図２）。観察者は、パターンが見えるか否かを判断する。階段法（Ｈ． Ｌｅｖｉｔｔ， Ｊｏｕｒｎａｌ ｏｆ ｔｈｅ Ａｃｏｕｓｔｉｃａｌ Ｓｏｃｉｅｔｙ ｏｆ Ａｍｅｒｉｃａ， Ｖｏｌ．４９， ｐ．４６７ （１９７１））のアルゴリズムにより、観察者が正答したら輝度コントラストを下げ、観察者が誤答したら輝度コントラストを上げる。このようにパターンの輝度コントラストを再度計算し、パターンと点滅光を再び提示する。この試行を繰り返すことにより、観察者がパターンを発見できる最小の輝度コントラスト値（コントラスト閾）を決定する。このコントラスト閾が大きいほど点滅光の存在により他のパターンが見えにくくなることを示しているため、この値を視覚的負荷量として定義する。そして、点滅光を構成する色を変えて、同様の測定を行う。その際には、二色の錐体応答量を様々な組み合わせを用いることにより、用いる出力装置で提示可能な色空間を全面的に調査する。この結果、コントラスト閾と錐体応答量は式（１）で表される関係にあり、先に記した重み付け係数値（α、β、γ）で実験結果を説明できることがわかった。すなわちＬＭ項（α）とＬｕｍｉｎａｎｃｅ項（γ）がコントラスト閾を増加させる強い効果を及ぼす一方、Ｓ項のコントラスト閾に対する効果（β）は、ＬＭ項やＬｕｍｉｎａｎｃｅ項の１／５である。また、点滅光の時間周波数特性は、時間周波数が約８Ｈｚ近辺で最大となるバンドパス型である（ｔ_ｐｅａｋ＝８、ｓｄ＝４）。すなわち、二色が８Ｈｚで点滅する時にコントラスト閾の上昇が最大となり、その時間周波数値から逸脱するにつれてコントラスト閾の変化量が小さくなる。
【００２３】
【発明の実施の形態】
次に、本発明の実施の形態について図面を参照して説明する。
【００２４】
図１を参照すると、本発明の一実施形態の視覚負荷度測定装置は色特性入力部１と点滅光作成部２と点滅光提示部３と点滅光測定部４と視覚負荷度測定部５と視覚負荷度出力部６で構成されている。
【００２５】
色特性入力部１は、点滅光を構成する二色の色特性と、点滅光の時間周波数をキーボード等から入力する。時間周波数は１秒間に何回色を切り替えるかを示す値である。入力された色特性に基づいて点滅光作成部２において、点滅光が作成される。点滅光はあらかじめ入力された二つの色を持つフレーム（画面）を、入力された時間周波数により、交互に提示するものである。この点滅光は、点滅光提示部３で画像提示装置（不図示）に提示される。色特性測定部４により、点滅光提示部３で提示された点滅光の色特性を測定する。そこで得られた色特性値を用いて、視覚負荷度計算部５において、点滅光を構成する二色の各々の色の錐体応答量（Ｌ_１、Ｍ_１、Ｓ_１、Ｌ_２、Ｍ_２、Ｓ_２）を計算し、これらの値を式（２）、（３）、（４）に代入してＬＭ、Ｓ、Ｌｕｍｉｎａｎｃｅを求め、別途算出したｆ（ｔ）（式（５））および重み付き係数α、β、γの値とともに式（１）に代入して視覚負荷度を計算する。なお、点滅光作成部２における作成、点滅光提示部３における提示、点滅光測定部４における測定を回避して、色特性入力部１に入力された色特性値を直接視覚負荷度計算部５へ送ることによっても、同様に視覚負荷度が計算できる。このようにして計算された視覚負荷度は、視覚負荷度出力部６によりディスプレイに表示され、またはプリンタで印刷されて、ユーザに提示される。
【００２６】
前者（図１）の装置と後者の装置（入力された色特性を視覚負荷度計算部５へ送る）との差は、画像提示装置の表示特性や画像提示装置がおかれている環境による影響を考慮に入れるか入れないかにあり、画像提示装置の表示特性や観測環境が重要な場合には、前者の測定装置が有効であり、色特性のみの負荷を測定するには、後者の装置で簡易に測定することができる。
【００２７】
後者の場合は、使用する画像提示装置におけるＲＧＢ信号のｘｙ色度座標が既知であれば、その値を用いて視覚負荷度を算出することができる。画像提示装置におけるＲＧＢ信号のｘｙ色度座標が未知である場合には、各画像提示装置において、一般的なＮＴＳＣ色度座標［（Ｘ_Ｒ＝０．６７，Ｙ_Ｒ＝０．３３），（Ｘ_Ｇ＝０．２１，Ｙ_Ｇ＝０．７１），（Ｘ_Ｂ＝０．１４，Ｙ_Ｂ＝０．０８）］を使うことができる。すると、色特性部１から入力した色パターンのＲＧＢ値（Ｌ_Ｒ，Ｌ_Ｇ，Ｌ_Ｂ）から三刺激値ＸＹＺが計算される（式（９））。この値を式（８）に代入することにより、錐体応答量を求めることができる。
【００２８】
この装置により、点滅光がどの程度の視覚負荷をもたらすのかを定量的に示すことができる。式（１）は、視覚負荷度が、点滅光を観察する観察者の網膜上における錐体の応答量に基づいていることを示している。式（１）からわかるように、点滅する色の間で、各色の反対色応答と輝度の違いが大きくなるにつれて、視覚負荷量が大きくなる。特にＬＭ反対色応答の差が視覚負荷度を大きくする。したがって、図３（１）に示したような色の組み合わせ（赤と青緑）による点滅光の視覚負荷度は大きい。視聴者に光過敏症を引きおこしたことで問題になったテレビ番組で用いられていた色の組み合わせは、ＬＭ反対色応答における色差が大きく、また点滅の頻度（時間周波数）が１２Ｈｚと最適値（８Ｈｚ）に近いため、１９．４という高い視覚負荷度を示した（図３（２））。
【００２９】
また、式（１）およびそのパラメータ値からわかるように、点滅光を構成する色により、点滅光を構成する二色の間でＳ反対色信号の応答量が変化しても、視覚負荷量は大きくは変化しない。このことから、Ｓ反対色信号の応答変化のみにより知覚される色による点滅光（例えば紫と黄緑）（図３（２））は、強い視覚負荷をもたらさない。実際上式に当てはめると、光過敏発作を引き起こす可能性のある赤と青の点滅光は、紫と黄緑の点滅光に比べて８倍以上視覚負荷度が高いことがわかった。このように、映像における視覚効果として色点滅光を用いる場合、今回の発明により、その視覚的負荷度を定量的に測定することができる。
【００３０】
推定されたパラメータ値からわかるように、点滅光を構成する二色間に輝度差がない場合、視覚負荷度はほぼＬＭ反対色信号の強さ（式（１）におけるＬＭ）に依存する。また、時間周波数の効果（式（１）におけるｆ（ｔ））は８Ｈｚを中心にした比較的広い範囲で大きく変化しない。したがって、輝度差がない場合には、式（１）のうちのαＬＭの項だけから視覚負荷度を推定することもできる。ただし、実際には二色間の輝度差をないという状況は稀であると考えられ、式（１）により全要因を考慮するほうが妥当である。
【００３１】
なお、図１に示した処理は専用のハードウェアにより実現されるもの以外に、その機能を実現するためのプログラムを、コンピュータ読み取り可能な記録媒体に記録して、この記録媒体に記録されたプログラムをコンピュータシステムに読み込ませ、実行するものであってもよい。コンピュータ読み取り可能な記録媒体とは、フロッピーディスク、光磁気ディスク、ＣＤ−ＲＯＭ等の記録媒体、コンピュータシステムに内蔵されるハードディスク装置等の記憶装置を指す。さらに、コンピュータ読み取り可能な記録媒体は、インターネットを介してプログラムを送信する場合のように、短時間の間、動的にプログラムを保持するもの（伝送媒体もしくは伝送波）、その場合のサーバとなるコンピュータシステム内部の揮発性メモリのように、一定時間プログラムを保持しているものも含む。
【００３２】
【発明の効果】
以上説明したように、本発明によれば、テレビや映画における視覚効果として用いられている点滅光の視覚負荷度を、人間の視覚生理学的メカニズムに基づいて推定することにより、視覚効果としての点滅光の視覚負荷度を適切に調節することが可能になる。この手法を用いれば、例えばビデオゲームやインターネットで配信されるコンテンツに含まれる視覚負荷が大きな刺激を選択的に排除したり、負荷が小さくなるように変調したりすることができる。
【図面の簡単な説明】
【図１】本発明の一実施形態の視覚負荷度測定装置のブロック図である。
【図２】式（１）を導き出す実験において用いた視覚パターンの例を示す図である。
【図３】視覚負荷度と色の組み合わせの関係例を示す図である。
【符号の説明】
１ 色特性入力部
２ 点滅光作成部
３ 点滅光提示部
４ 点滅光測定部
５ 視覚負荷度計算部
６ 視覚負荷度出力部[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method and apparatus for measuring the visual load of blinking color light.
[0002]
[Prior art]
The flashing light created by preparing two uniform single-colored backgrounds and presenting them quickly and repeatedly has been frequently used as a visual effect in television and movie images. However, theoretical studies on the selection of colors that make up the blinking light have not been advanced. Due to the lack of such knowledge, the use of blinking light in a combination of red and blue as a visual effect of television has caused many viewers to have light-sensitive seizures and become a social problem (1997). Thereafter, on April 8, 1998, "Guidelines on Video Techniques such as Animation" were formulated (Non-Patent Document 1). The guidelines call for the following three points that require careful attention:
[0003]
1. Flashing of images and light, especially flashing of “bright red” 2. Inversion of a scene with high contrast or sudden scene change. Use of regular patterns
[Non-patent document 1]
Commentary on "Guidelines on Video Techniques such as Animation", June 1998, Japan Broadcasting Corporation [0005]
[Problems to be solved by the invention]
The flickering light did not define the visual load felt by the observer. Therefore, no definition has been established on how to select the color that produces the flashing light. Also, the relationship between the choice of color in the blinking light and the resulting visual load and the neurophysiological mechanisms in the human visual system was unknown.
[0006]
An object of the present invention is to provide a method, an apparatus, a program, and a recording medium for measuring a visual load, which is a visual load felt by an observer with respect to color blinking light.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, the visual load degree measuring method of the present invention includes a step of inputting a color characteristic of a color constituting a blinking light, a time frequency of the blinking light, and an input color characteristic and a time frequency. Creating the blinking light based on the generated blinking light; measuring the color characteristics of the presented blinking light; and configuring the blinking light using the measured color characteristics. The method comprises the steps of calculating a cone response amount of each color, calculating a visual load from the response amount and the time frequency, and outputting the calculated visual load.
[0008]
Hereinafter, the principle of the present invention will be described.
[0009]
A measure of the visual load is how much the blinking light impairs the visibility of other simultaneously presented patterns. That is, it is assumed that blinking light that makes other patterns difficult to see has a high visual load. From the results of the experiments described later, it was found that the visibility of other patterns greatly changed depending on the combination of colors constituting the blinking light, the luminance difference between the colors, and the time frequency of blinking. One type of color flashing light greatly reduces the visibility of the pattern presented over it, while another type of color flashing light does not significantly obstruct the pattern visibility. Therefore, it is assumed that the former has a higher visual load of the color blinking light than the latter. From the results of the experiment, it was found that the visual load was expressed by the following equation in the blinking light of two colors. It is assumed that the greater the value of Expression (1), the higher the visual load.
[0010]
(Equation 1)

Here [0011]
(Equation 2)

It is.
[0012]
LM in equation (1) is the difference between the two colors in the LM opposite color signal represented by equation (2), S in equation (1) is the difference between the two colors in the S opposite color signal represented by equation (3), Luminance in equation (1) is the difference in luminance between the two colors represented by equation (4). Details will be described later. α, β, and γ indicate the weighting coefficients of the terms LM, S, and Luminance, respectively. In the formulas (2), (3), and (4), L ₁ , M ₁ , and S ₁ are the response amounts of the L cone, the M cone, and the S cone provided by one of the colors constituting the blinking light, and L ₂ , M ₂ , and S ₂ are the response amounts of the L cone, M cone, and S cone provided by the other color. F (t) is a function indicating how the effect provided by the color difference and the luminance difference between the two colors changes depending on the time frequency of the blinking light (the number of times of blinking per second). Is represented by Here, t is the flashing light time frequency (in Hz), t _peak time frequency effect of flashing light has a peak, sd is the standard deviation which indicates the spread of the function. The weighting coefficients (α, β, γ) and the effect of the time frequency (f (t)) were obtained by experiments described later. As a result, the weighting coefficient is α = 1
β = 0.2
γ = 1
And two parameters in the time-frequency function f (t) are t _peak = 8, respectively.
sd = 4
It becomes.
[0013]
In the human retina, three types of photoreceptor cells (L, M, and S cones) called cones are distributed. The cone response amount is an amount that activates cones present in the retina of the observer when the observer views the visual pattern. Since each of the three cones has different spectral characteristics, it receives a unique response to a light stimulus. The responses from these three pyramids are the basis of our color vision.
[0014]
The response amount of each cone can be obtained from the spectral sensitivity characteristics of each cone (a sensitivity function for each wavelength of each cone) and the spectral radiance of the color presented on the screen. Since there is no great individual difference in the spectral sensitivity characteristics of each cone, an average value generally called a basic sensitivity function is used (VC Smith, J. Pokorney, Vision Research, Vo. 15, pp. 161-161). 171 (1975)). The response amount (L, M, S) of the cone is represented by the following equation (6).
[0015]
[Equation 3]

Here, l (λ), m (λ) and s (λ) represent the basic sensitivity function of each cone, and C (λ) represents the spectral radiance of the color presented on the screen. Even when the spectral radiance of the color presented on the screen cannot be measured, the xy chromaticity coordinate value (x, y) and the luminance value (Lum) of the color are measured (or before the color is presented on the screen). If predetermined), the cone response amount can be calculated. First, the xy chromaticity coordinate value (x, y) and the luminance value (Lum) are converted into tristimulus values X, Y, Z in the XYZ color system.
[0016]
(Equation 4)

Then, the response amount of the cone can be obtained from the tristimulus values X, Y, and Z by the following equation (8).
[0017]
(Equation 5)

Here, k is a constant for normalization, and _{(x l, y l, z} l) (x m, y m, z m) (x s, y s, z s) respectively L, M, S It is a convergence point of a confusion line in the xy chromaticity coordinates of each cone, and means an xy chromaticity value that uniquely identifies each of the L, M, and S cones. These values are described in Smith & Pokory (1975).
[0018]
Further, even when the color to be presented on the screen is defined in advance by RGB values, the response amount of the cone can be calculated. First, tristimulus values X, Y, and Z are obtained by the following equation (9).
[0019]
(Equation 6)

_{(X R, y R, z} R), (x G, y G, z G), (x B, y B, z B) of R in the image display device to be used, G, xy chromaticity coordinates of the B signal Value. Since these values differ depending on the device used, it is necessary to measure them in advance. _{(L R, L G, L} B) are, R bring each color flashing light, the luminance value of the G, B signals. By substituting the tristimulus values X, Y, and Z obtained from Expression (9) into Expression (8), the response amount of the trigonal cone (L, M, S) can be obtained.
[0020]
As described above, the cone response amounts (L ₁ , M ₁ , S ₁ , L ₂ , M ₂ , S ₂ ) are based on various characteristics (flashing light) of each color constituting the flashing light measured by the color characteristic measuring unit. It can be calculated from the xy chromaticity coordinate value of each color, the luminance value of each color of blinking light, and the spectral sensitivity characteristic of each color of blinking light) and the known spectral sensitivity characteristics of three cones (Smith & Pokory, 1975). Alternatively, in the color characteristic input unit, the cone response amount can be calculated from color characteristic values (xy chromaticity coordinate values of RGB signals of the image presentation device, and luminance values of RGB of each color of blinking light) input in advance. .
[0021]
If the response amounts (L ₁ , M ₁ , S ₁ , L ₂ , M ₂ , S ₂ ) that the two colors constituting the flashing light bring to the three cones and the time frequency (t) of the flashing light are known, the equation ( The visual load degree can be calculated from 1). Equation (1) has terms indicating the color difference and the luminance difference between the two colors of the blinking light. The LM term in the equation (1) is a difference of a color signal (difference in response between the L cone and the M cone) called an LM opposite color carried by a group of nerve cells called the Parvo cell pathway (De Valois & De Valois, 2000). , Seeing, Academic Press). LM opposite color components of one color constituting the flashing light _{(L 1 -M 1) / (} L 1 + M 1), LM opposite color component of the other color _{(L 2 -M 2) / (} L 2 + M ₂ ). In the LM term of the equation (1), the absolute value of the difference between the LM opposite color components is calculated. The S term in the equation (1) is a difference in a color signal (a difference between the sum of the responses of the L cone and the M cone and the response of the S cone) called an S opposite color carried by a group of nerve cells called the Konio cell pathway. is there. The S opposite color component of one of the colors constituting the blinking light is S ₁ − (L ₁ + M ₁ ), and the S opposite color component of the other color is S ₂ − (L ₂ + M ₂ ). In the S term of the equation (1), the absolute value of the difference between the S opposite color components is calculated. The Luminance term in Expression (1) is a difference between luminance signals (sum of responses of the L cone and the M cone) carried by a group of nerve cells called the Magno cell pathway. The luminance component of _one color of the blinking light is L ₁ + M ₁ , and the luminance component of the other color is L ₂ + M ₂ . In the Luminance term of Expression (1), the absolute value of the difference between the luminance components is calculated.
[0022]
How these three components (LM, S, Luminance) affect the visual load degree (α, β, γ), and how the frequency of blinking light (time frequency t) The influence was derived based on the following psychological experiment. First, a pattern having a certain brightness contrast (sinusoidal lattice fringe pattern) is created by a computer. Next, two colors of the background light constituting the blinking light are selected, and the pattern and the blinking light are simultaneously presented to the observer (FIG. 2). The observer determines whether the pattern is visible. According to the algorithm of the staircase method (H. Levitt, Journal of the Acoustic Society of America, Vol. 49, p. 467 (1971)), if the observer answers correctly, the luminance contrast is lowered, and if the observer makes an incorrect answer, the luminance contrast is increased. . Thus, the brightness contrast of the pattern is calculated again, and the pattern and the blinking light are presented again. By repeating this trial, the minimum luminance contrast value (contrast threshold) at which the observer can find the pattern is determined. This value is defined as a visual load because the larger the contrast threshold is, the more difficult it is to see other patterns due to the presence of blinking light. Then, the same measurement is performed while changing the color of the blinking light. In this case, the color space that can be presented by the output device used is thoroughly investigated by using various combinations of the cone response amounts of the two colors. As a result, it has been found that the contrast threshold and the cone response amount have a relationship represented by Expression (1), and the experimental results can be explained by the weighting coefficient values (α, β, γ) described above. That is, while the LM term (α) and the Luminance term (γ) have a strong effect of increasing the contrast threshold, the effect (β) of the S term on the contrast threshold is １ ／ of the LM term and the Luminance term. The time-frequency characteristic of the blinking light is a band-pass type in which the time frequency becomes maximum around about 8 Hz (t _peak = 8, sd = 4). That is, when the two colors blink at 8 Hz, the rise of the contrast threshold becomes maximum, and the amount of change in the contrast threshold becomes smaller as the time deviates from the time frequency value.
[0023]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings.
[0024]
Referring to FIG. 1, a visual load degree measuring device according to an embodiment of the present invention includes a color characteristic input unit 1, a blinking light creating unit 2, a blinking light presenting unit 3, a blinking light measuring unit 4, a visual load degree measuring unit 5, It comprises a visual load degree output unit 6.
[0025]
The color characteristic input unit 1 inputs, from a keyboard or the like, the color characteristics of the two colors constituting the blinking light and the time frequency of the blinking light. The time frequency is a value indicating how many times a color is switched per second. A blinking light is created in the blinking light creating unit 2 based on the input color characteristics. The blinking light alternately presents frames (screens) having two colors that have been input in advance according to the input time frequency. This blinking light is presented to an image presentation device (not shown) by the blinking light presentation unit 3. The color characteristics of the blinking light presented by the blinking light presentation unit 3 are measured by the color characteristic measuring unit 4. Using the obtained color characteristic values, the visual load degree calculator 5 calculates the cone response amounts (L ₁ , M ₁ , S ₁ , L ₂ , M ₂₎ of each of the two colors constituting the blinking light. , S ₂ ), and substitute these values into equations (2), (3), and (4) to determine LM, S, and Luminance, and calculate f (t) (equation (5)) and f (t) separately calculated. The visual load is calculated by substituting the values of the weighted coefficients α, β, and γ into the equation (1). Note that the color characteristic value input to the color characteristic input unit 1 is directly converted to the visual load calculation unit 5 by avoiding creation by the blinking light generation unit 2, presentation by the blinking light presentation unit 3, and measurement by the blinking light measurement unit 4. The visual load degree can be calculated in the same manner by sending it to. The visual load degree calculated in this way is displayed on a display by the visual load degree output unit 6 or printed by a printer and presented to the user.
[0026]
The difference between the former device (FIG. 1) and the latter device (sending the input color characteristics to the visual load calculation unit 5) is affected by the display characteristics of the image presentation device and the environment in which the image presentation device is placed. If the display characteristics of the image presentation device and the observation environment are important, the former measurement device is effective.To measure the load of only the color characteristics, use the latter device. It can be easily measured.
[0027]
In the latter case, if the xy chromaticity coordinates of the RGB signals in the image presentation device to be used are known, the visual load can be calculated using the values. If xy chromaticity coordinates of the RGB signal in the image display device is unknown, in each image display device, a general NTSC chromaticity coordinates _{[(X R = 0.67, Y} R = 0.33), ( _{X G = 0.21, Y G =} 0.71), it can be used _{(X B = 0.14, Y B} = 0.08)]. Then, RGB values of the color pattern inputted from the color characteristic unit _{1 (L R, L G,} L B) tristimulus values XYZ from are calculated (equation (9)). By substituting this value into equation (8), the cone response amount can be obtained.
[0028]
With this device, it is possible to quantitatively indicate the degree of visual load caused by the blinking light. Equation (1) indicates that the visual load is based on the response amount of the cone on the retina of the observer observing the blinking light. As can be seen from equation (1), the visual load increases as the difference between the opposite color response and the luminance of each blinking color increases. In particular, the difference in LM opponent color response increases the visual load. Therefore, the visual load of the blinking light by the combination of colors (red and blue-green) as shown in FIG. The color combination used in the television program that caused the photosensitivity to the viewer had a large color difference in the LM opposite color response, and the blinking frequency (time frequency) was 12 Hz, which was the optimal value. (8 Hz), it exhibited a high visual load of 19.4 (FIG. 3 (2)).
[0029]
Further, as can be seen from the equation (1) and the parameter values, even if the response amount of the S opposite color signal changes between the two colors constituting the blinking light, the visual load amount varies depending on the color constituting the blinking light. It does not change much. For this reason, the blinking light (for example, purple and yellow-green) due to the color perceived only by the response change of the S opposite color signal does not cause a strong visual load. In fact, when applied to the equation, it was found that the flashing light of red and blue, which may cause a photosensitivity attack, is more than eight times as high as the flashing light of purple and yellow-green. As described above, when color blinking light is used as a visual effect in an image, the visual load can be quantitatively measured by the present invention.
[0030]
As can be seen from the estimated parameter values, when there is no luminance difference between the two colors that make up the blinking light, the visual load largely depends on the strength of the LM opposite color signal (LM in equation (1)). Further, the effect of the time frequency (f (t) in the equation (1)) does not largely change in a relatively wide range centered on 8 Hz. Therefore, when there is no luminance difference, the visual load can be estimated only from the term of αLM in Expression (1). However, it is considered that there is rarely a situation in which there is no luminance difference between two colors, and it is more appropriate to consider all the factors by the equation (1).
[0031]
It should be noted that the processing shown in FIG. 1 is not only realized by dedicated hardware, but also a program for realizing the function is recorded on a computer-readable recording medium, and the program recorded on this recording medium is recorded. May be read into a computer system and executed. The computer-readable recording medium refers to a recording medium such as a floppy disk, a magneto-optical disk, a CD-ROM, or a storage device such as a hard disk device built in a computer system. Further, the computer-readable recording medium is one that dynamically holds a program for a short time (transmission medium or transmission wave), such as a case where the program is transmitted via the Internet, and serves as a server in that case. It also includes those that hold programs for a certain period of time, such as volatile memory inside a computer system.
[0032]
【The invention's effect】
As described above, according to the present invention, the visual load of the blinking light used as a visual effect in a television or a movie is estimated based on a human visual physiological mechanism, so that the blinking as a visual effect is performed. It is possible to appropriately adjust the visual load of light. By using this method, for example, a stimulus having a large visual load included in a video game or content distributed on the Internet can be selectively excluded or modulated so as to reduce the load.
[Brief description of the drawings]
FIG. 1 is a block diagram of a visual load measurement device according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating an example of a visual pattern used in an experiment for deriving Expression (1).
FIG. 3 is a diagram illustrating an example of a relationship between a visual load and a combination of colors.
[Explanation of symbols]
1 color characteristic input unit 2 blinking light creation unit 3 blinking light presentation unit 4 blinking light measurement unit 5 visual load degree calculation unit 6 visual load degree output unit

Claims (6)

Inputting a color characteristic of a color constituting the blinking light and a time frequency of the blinking light;
Creating a blinking light based on the input color characteristics and the time frequency;
Presenting the created flashing light;
Measuring the color characteristics of the presented flashing light;
Using the measured color characteristics, calculating the cone response amount of each color constituting the blinking light, calculating the visual load degree from the response amount and the time frequency,
A method for measuring a visual load, comprising the step of outputting the calculated visual load. Inputting the color characteristics of the colors that make up the flashing light;
Using the input color characteristics, calculating the cone response amount of each color constituting the blinking light, calculating the visual load degree from the response amount and the time frequency,
A method for measuring a visual load, comprising the step of outputting the calculated visual load. Means for inputting a color characteristic of a color constituting the blinking light and a time frequency of the blinking light,
Means for creating blinking light based on the input color characteristics and time frequency,
Means for presenting the created flashing light,
Means for measuring the color characteristics of the presented flashing light;
Using the measured color characteristics, calculating the cone response amount of each color constituting the blinking light, means for calculating the visual load from the response amount and the time frequency,
A visual load degree measuring device having means for outputting the calculated visual load degree. Means for inputting the color characteristics of the color comprising the blinking light;
Using the input color characteristics, calculate the cone response amount of each color constituting the blinking light, means to calculate the visual load from the response amount and the time frequency,
A visual load degree measuring device having means for outputting the calculated visual load degree. A visual load degree measuring program for causing a computer to execute the visual load degree measuring method according to claim 1. A computer-readable recording medium that records a visual load degree measurement program for causing a computer to execute the visual load degree measurement method according to claim 1.

Images