Skip to content

Instantly share code, notes, and snippets.

@liangtai
Last active August 29, 2015 13:57
Show Gist options
  • Save liangtai/9617591 to your computer and use it in GitHub Desktop.
Save liangtai/9617591 to your computer and use it in GitHub Desktop.
Elle Stone 氏による GIMP ヘルプ文書の色管理の章の改訂草稿
<!DOCTYPE HTML>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<link rel="shortcut icon" href="favicon.ico" type="image/x-icon"/>
<title>第11章 GIMP の色管理 (改訂草稿)</title>
</head>
<style type="text/css">
.clear {
overflow : hidden;
clear : both;
display : block;
visibility : hidden;
width : 0;
height : 0;
font-size: 0;
line-height: 0;
margin: 0;
}
html {
background-color: #888;
width: 100%;
}
body {
font-size: 100%;
background-color: #fafafa;
max-width: 52rem;
padding: 0 4rem;
margin: 0 auto; /* this centers body*/
height: 100%;
min-height: 100%;
line-height: 2;
font-family: "DejaVu Sans", "Lucida Grande", "Lucida Sans Unicode", Verdana, sans-serif;
}
figure {
max-width: 45rem;
margin: 0;
border: 1px solid #000;
padding: 2rem;
background: #474747;
color: #fff;
margin: 2rem 0;
}
figure a {color: #aaf;}
#gimppad {
padding: 1rem;
}
img {max-width: 100%;display: block; position: relative; padding: 1rem;}
h2 {text-align: left;}
#gimppad {
counter-reset: h2counter h3counter h4counter h5counter formulacounter figurecounter tablecounter;
}
#gimppad h2:before {
content: counter(h2counter,upper-alpha) ".\0000a0";
counter-increment: h2counter;
}
#gimpad h2 {counter-reset: h3counter;}
</style>
<body>
<div id="gimppad">
<P>
このページは Elle Stone 氏が GIMP ヘルプ文書チームに提案した
「<a href="http://docs.gimp.org/2.8/ja/gimp-imaging-color-management.html">第11章 GIMP の色管理</a>」
の改訂版 (原文は
<A href="http://ninedegreesbelow.com/gimpgit/gimpdoc/gimp-doc-chapter-11.html"
>Proposed Replacement for Chapter 11. GIMP Color Management</A>)
の日本語対訳です。
Elle Stone 氏はリンク先ページの文章および画像の著作権を主張しており、
その内容が GIMP ヘルプ文書の一部となった時に適切なライセンスで公開しなおすつもりだそうです。
この対訳ページのソースが
<A HREF="https://gist.github.com/liangtai/9617591">GIST.GITHUB</A>
でも入手できます。
</P>
<li>Sections A and B are complete except for any revisions.</li>
<li>Sections C and D each need one additional subsection to be written.</li>
<li>Sections E, F, and G are currently just placeholders.</li>
<h1>Chapter 11. GIMP Color Management<br/>
11. GIMP の色管理</H1>
<p class="bold">Table of Contents<br/>
目次</p>
<ol style="list-style-type: upper-alpha;">
<li><a href="#introduction">Introduction to color management<br/>
色管理のあらまし</a></LI>
<li><a href="#monitor">Calibrating and profiling your monitor<br/>
モニターのキャリブレーションとプロファイリングをしよう</a></LI>
<li><a href="#color">What is color?<br/>
色とは何か</a></LI>
<li><a href="#color-gamuts">Color gamuts<br/>
色ガモット</a></LI>
<li><a href="#proofing">Soft proofing<br/>
ソフトプルーフ</a></LI>
<li><a href="#gimp-specific">GIMP-specific information<br/>
GIMP のみに関わる話題</a>
<ol>
<li>open/assign/convert</li>
<li>unbounded mode high bit depth color conversions</a></li>
<li>global and per image color management and soft proofing options</li>
</ol>
</li>
<li><a href="#more-information">Links to more information</a></li>
</ol>
<A id="introduction"></A>
<h2>Introduction to color management<br/>
色管理のあらまし</H2>
<h3>What color management is<br/>
色管理 (カラーマネジメント) とは何か</H3>
<p>Devices that produce or display color include:<br/>
色彩表現や発色表示を行なうデバイスはつぎに挙げるようにさまざまなものがあります。
</P>
<ul>
<li>scanners and cameras<br/>
スキャナーやカメラ</LI>
<li>computer monitors and other digital display devices (tablets, digital picture frames, HDR displays, and so on)<br/>
コンピューターのディスプレイモニターのようなディジタル表示デバイス
(タブレット、 ディジタル額縁、 HDR ディスプレイ、 などなど)</LI>
<li>printers &mdash; for printers, the actual "device" is the printer plus the inks plus the paper<br/>
プリンター &mdash; 厳密にはプリンター機及びインク及び紙からなるデバイス</LI>
</ul>
<p>Broadly speaking, <dfn>color management</dfn> refers to any method of ensuring that color produced or displayed by one device can be <em>reliably reproduced</em> by another device. Note the use of the word "reliably" rather than "exactly": <a href="#color-gamuts">color gamut differences between devices</a> often prevent colors produced by one device from being exactly reproduced on another device.<br/>
大雑把な言い方をすれば
何かのデバイスで出力・表示した色が他のデバイスで <em>確実に再現</em>
できるようにするどんな手段を執っても <dfn>色管理</dfn> を行ったことになります。
ここで「完璧に exactly」ではなく「確実に reliably」の語を用いたのには理由があります。
<a href="#color-gamuts">デバイス間で色ガモットが異なる</a> ために、
とあるデバイスで出力した色が他のデバイスでは全く同じ色に再現できないという場合がよくあるからです。
</P>
<h3>Why color management is important<br/>
色管理が重要な理由</H3>
<p>Typically images are produced and displayed on a whole series of devices. For example:<br/>
画像の作成から表示まで多種多用なデバイスが利用される典型的な例を見てみましょう。
</P>
<ul>
<li>A digital photograph might be displayed on a monitor screen for color correction, and later be printed and hung on a wall, displayed in a digital photo frame, and sent to a magazine for mass reproduction. <br/>
ディジタル写真を撮ったあと、
色の調整をするならモニター画面で表示するでしょうし、
プリントして壁に架けたり、 ディジタルフォトフレームに表示させたり、
雑誌に投稿すれば大量の印刷物となることも考えられます。
</LI>
<li>A color slide from a film camera might be scanned in as digital image, color corrected on a monitor screen, and then made into a paper print and also displayed on the web.<br/>
フィルムカメラで撮ったカラースライド映写機用の画像を、
スキャンしてディジタル画像化し、 色を直すためにモニター画面に映し出し、
そのあと紙に印刷したりネット上に公開して閲覧することもあるでしょう。
</LI>
</UL>
<p><b>Color reproduction varies greatly from one device to another.</b> So achieving consistent color across different devices requires some form of color management. Without color management:<br/>
<B>デバイスが替わるたび色の再現は大幅に変化します。</B>
そのため異なるデバイス間で色に一貫性をもたせるには何らかの色管理のしくみが必要になるのです。
色管理ができていない場合はこんなことが起きます。</P>
<ul>
<li>Different cameras photographing the same scene at the same time will produce different colors, depending on the camera sensor, settings, and internal image processing (for digital cameras).<br/>
同じ被写体を同時に撮った写真でも使ったカメラが異なれば、 (ディジタルカメラの)
センサー、 設定、 内部と外部での画像処理方法の違いによって色に差異が生じます。
</LI>
<li>When displayed on different monitors, the same digital image will look very different, depending on the monitor's color-producing technology, and internal and external calibration settings.<br/>
表示させるモニターが異なれば、 ディスプレイモニターの色再現技術の違いや、
内部と外部のキャリブレーション設定の違いによって、
同一のディジタル画像がかなり異なって見えます。
</LI>
<li>When the same digital image is sent to different printers, then depending on the inks, paper, printer, and printer settings, the resulting print colors will be different.<br/>
同一のディジタル画像を別々のプリンターに送ったところ、
インクや紙質やプリンター機種やプリンターの設定の違いによって、
刷り上がりの色が異なって出てきます。
</LI>
</UL>
<h3>Managing color by device calibration<br/>
デバイスのキャリブレーションによる色管理の方法
</H3>
<p><b>Device calibration means <em>altering</em> how a device produces color, the goal being to put the device in a known "target" state of color reproduction.</b> Before ICC profiles came on the scene, the main approach to achieving consistent color reproduction between devices was by device calibration. The big print houses, for example, would calibrate their printers to produce color that was acceptable to their customers, and then they would calibrate their monitors to match the colors produced by their printers. <br/>
<B>
ここで言うデバイスのキャリブレーションとは、 色再現出力結果に「標準」を定めて、
デバイスがその状態に近づくよう発色機能を <em>補正</em> することを指します。
</B>
ICC プロファイルが登場する以前、
デバイス同士で色の再現に一貫性を保たせるときの方法といえばデバイスのキャリブレーションが主流でした。
たとえば大きな印刷所では、 顧客の納得が得られなくてはいけませんから
工場内の印刷機をすべてキャリブレートして発色を補正し、
さらに印刷機と同じ色が表示てきるようにモニターもキャリブレートしました。
</P>
<h3>ICC profile color management<br/>
ICC プロファイルによる色管理</H3>
<p>There are two problems with color management by device calibration:<br/>
デバイスのキャリブレーションによる色管理は問題を 2 つ抱えています。</P>
<ol>
<li>As soon as you share digital images with someone who's not using your in-house calibrated devices, consistent color reproduction is lost.<br/>
お持ちのデバイス同士でキャリブレーションを行なってもそれ以外のデバイスを使っている誰かとディジタル画像を共有すればたちまち色の再現の一貫性が失われます。</LI>
<li>Devices vary in the degree to which they can be calibrated. Monitor displays and printers can (and should) be calibrated, though sadly not all monitors and printers provide adequate calibration controls. Scanners and cameras <em>can't be calibrated</em> &mdash; calibrating a camera or sensor would require the ability to change how the sensor responds to light.<br/>
キャリブレーションで補正が可能な範囲はデバイスによってかなり異なります。
ディスプレイモニターやプリンターはキャリブレーションが可能ですしそうでなくてはいけませんが、
残念ながらあらゆるモニターやプリンターが存分にキャリブレーションの制御を許しているわけではありません。
スキャナーとカメラは <em>キャリブレートができません。</em>
カメラやスキャナーは
光学読み取り装置の反応を変更できなければキャリブレーションできないからです。
</LI>
</OL>
<p><dfn>ICC profile color management</dfn> solves both of these problems. Whereas device calibration <em>alters</em> a device's color reproduction characteristics, an ICC profile <em>describes</em> a device's color reproduction characteristics. ICC profile information is stored in a specific format described in detail by the <a href="">ICC profile specifications</a>. An ICC profile can either be embedded in an image or saved to disk as an independent file.<br/>
<DFN>ICC プロファイルを用いた色管理</DFN> はその両方の問題を解決します。
デバイスのキャリブレーションがデバイスごとの色再現の特性を <em>補正</em> する一方で、
ICC プロファイルはデバイスごとの色再現の特性を <em>記述</em> します。
ICC プロファイル情報は <a href="">ICC プロファイル仕様</a>
に詳しく定められた形式で保存されます。
ICC プロファイルは画像に情報を埋め込むか別途ファイルに記録する方法のどちらも可能です。
</P>
<p>ICC profiles are created by a process called <dfn>profiling</dfn>. The procedure for profiling a device depends on the type of device being profiled:<br/>
ICC プロファイルは <DFN>プロファイリング</DFN> という工程を経て作成します。
プロファイリングされるデバイスの種類別に作成の工程を分類するとつぎのようになります。
</P>
<ul>
<li>Cameras and scanners are profiled by photographing or scanning paper targets printed with patches of known colors, and then associating the RGB values of the resulting scan or photograph with the known colors in the target. <i>Note: Camera profiling works best if you shoot raw</i> &mdash; otherwise you are trying to undo the unpredictable results of in-camera-applied picture styles.<br/>
カメラやスキャナーのプロファイルは、
規定の標本色を印刷した紙を撮影/スキャンして、
出力される写真やスキャンデータと色標本紙の各色の RGB 値を比較する方法で得ます。
<I>ここでご注意。 カメラをプロファイリングするときはロウモードが適しています。</I>
他のモードで撮るとカメラ内で写真が加工されてしまうので、
想定外な出力を除けて元の情報を取り戻さなくてはならないはめに陥るからです。
</LI>
<li>RGB monitors are profiled by sending various combinations of intensities of Red, Green, and Blue to the screen, and measuring the resulting color.<br/>
RGB モニターのプロファイルをするときは、
赤・緑・青の強度を変化させ幾通りも組み合わせた色の信号を画面に送り、
表示されるそれぞれの色を測定して行います。
</LI>
<li>A printer-paper-ink combination is profiled by telling the printer to print patches of varying combinations of inks on the paper, and then measuring the color of each patch. <br/>
プリンターについては、
プリンターと用紙とインクの組み合わせごとにプロファイリングを行います。
インクの配合を変えたさまざまな色標本を紙面に出力させ、
刷り出された色をそれぞれ測定します。
</LI>
</UL>
<p>ICC profile color management was developed in the 1990s by the <a href="http://color.org">International Color Consortium</a> (hence "ICC"), which has the stated goal of "promot[ing] the use and adoption of open, vendor-neutral, cross-platform color management systems". Initially ICC profile color management was mostly confined to the big print houses that could afford to buy computers that were fast enough to use for digital color correction. But as computers got faster and more photographers and artists started working with digital images, GIMP and other image editing programs began including ICC profile color management as a matter of course.<br/>
ICC プロファイルによる色管理のしくみは 1990 年代に登場しました。
これを開発したのはインターナショナル カラー コンソーシアム (ICC:
<A HREF="http://color.org">International Color Consortium</A>) という、
「企業間の壁を越え、 各機種共通のオープンな色管理システムの利用と適合を促進する」
ことを使命とする業界団体です。
最初期に ICC プロファイルを採用したのはディジタル色補正を実行するのに十分な処理速度をもつコンピューターが導入できる資金力のある大手印刷所がほとんどでした。
しかしコンピューターがどんどん速くなり写真家や芸術家がディジタル画像を扱いはじめると、
GIMP をはじめ画像加工プログラムも当然のように ICC プロファイルによる色管理をとり入れはじめます。
</P>
<figure>
<figcaption>Figure 11.1. <b>Viewing and editing images with and without color management</b><br/>
図 11.1.
<B>色管理ができていない状態で画像を表示し加工した場合</B></figcaption>
<p>The images below illustrate why the only way to get consistent, predictable color from one color reproduction device to the next is to use a color-managed workflow:<br/>
つぎの図は色管理をとり入れた作業工程だけが、
一方のデバイスが出力する色と他の色再現デバイスの表示色がつねに同様で予測可能なものとなる唯一の方法であることを示しています。
</P>
<ol>
<li><p>Typical color corrections to a digital image include color balancing (so neutral colors really are neutral), and modifying contrast and saturation (to add more "pop" to the image). Our sample original image (#1 below) doesn't need to be color balanced because the neutral colors &mdash;the snow and clouds &mdash; are already neutral enough (that is, slightly on the blue side of white, suitable for a cold winter scene). But the shadows could be darker, the highlights could be brighter, and a little added color saturation wouldn't hurt.<br/>
ディジタル画像の色の修正作業のなかでも、
色のバランス補正 (つまり灰色系統を実際に無色化) だとか、
コントラストと彩度の調節 (もっと目を魅くところを画像にプラス)
などはわかりやすい典型的な直し方です。
そこで実例画像 (例図の 1 番目) を用意しましたがこれに色のバランス補正は不要です。
色がない「雪や雲」が既に十分に無色なためです
(これが僅かに青みがかった白であれば冷えきった冬の光景にぴったりです)。
一方で影の部分はもっと暗く、 ハイライト部分はもっと明るくした方が良さそうですし、
彩度を少し加えてみても悪化はしないようです。
</P>
<p><img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-original.jpg" style="float: left;">Image #1 is an RGB image, which means the actual image colors are determined by the image's RGB values plus its embedded or assigned ICC profile. In a color-managed workflow, the colors you see on your calibrated and profiled monitor are a close match to the image's actual colors. <br/>
1 番目の例図は RGB 画像なので、
つまりこの画像の実際の色は画像の RGB 値と、
画像に埋め込まれているか関連づけられている ICC プロファイルから算定されることになります。
色管理がなされている作業工程において、
キャリブレーションされプロファイリングされたモニターで見えている色は
画像の実際の色にぴったり合っています。
</P>
<p>Without color management and depending on the actual monitor, an uncalibrated monitor might make Image #1 <em>look</em> more magenta or blue or yellow than it really is. When attempting to edit an image displayed on such a monitor, the natural temptation is to add the <em>opposite</em> color to neutralize the monitor-produced "faux" color cast. For example, the temptation is to add green to images displayed on a monitor that makes images look more magenta than they really are.<br/>
色管理がなされずにモニターに頼ってしまうと、
キャリブレーションしていないモニターで 1 番目の画像が <EM>見掛け上</EM>
元の実際の色よりもマゼンタがかったり青みがかったりイエローっぽくなったように感じられることもありえます。
画像をそのような状態のモニターに表示して編集に臨んだ場合、
モニターが出力する偏った色調を正常化すべく
<EM>対向</EM> 色を加えたくなる自然な誘惑に駆られるはずです。
例えば常識的色彩感覚が画像に緑色の追加を求めているとき、
実際はモニターが元の色よりもマゼンタがかって見えるように表示していたからかもしれないのです。
</P>
<p>Below are side-by-side illustrations of viewing, editing, and printing our sample original image on a sample LCD monitor, with and without using color management:<br/>
つぎに示す図は色管理を行なった場合と行なっていない場合の表示・編集・印刷の違いを先に挙げた実例図を使って説明しています。
</P>
</LI>
<li>
<p>#2A below is what the original image looks like when displayed on a properly calibrated and profiled monitor in a color-managed workflow. #2B below shows what the exact same original image might look like when displayed on an uncalibrated monitor that displays a "faux" magenta color cast. This particular uncalibrated monitor also slightly crushes image's shadows:<br/>
下図 2 番目の A は色管理ができている工程で適切にキャリブレートされているモニターに実例図を表示した様子です。
同じ下図 2 番目の隣の B はキャリブレートせずマゼンタに偏ったモニターで全く同じ画像を表示した様子です。
ご覧のとおりキャリブレートしていないモニターは画像の影の部分にもややつぶれが見うけられます。
</P>
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-2A.jpg" style="float: left;">
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-2B.jpg" style="float: left;">
</li><div class="clear"></div>
<li>
<p>It's actually possible to edit the original image (#1 above) to <em>look</em> very similar on both the properly profiled and calibrated monitor (see #3A below) and the example uncalibrated, unprofiled monitor (see #3B below).<br/>
元の画像 (上の 1 番目) の表示画面が適切にプロファイルされキャリブレートできているモニターのもの (下図 3 番目の A の画像) なのかキャリブレートもプロファイルもなされていないモニターのもの (下図 3 番目の B の画像) なのか、
<EM>見掛け上</EM> 非常に似通っていたために見分けがつかないまま画像を編集することは実際にありうることです。
</P>
<p>Both images below have been given added saturation and contrast. For image #3B, <em>green</em> was added to make it <em>look</em> neutral on the uncalibrated, unprofiled monitor. Also the shadows in image #3B were lightened to make them <em>look</em> less crushed (they weren't actually crushed to begin with):<br/>
下図の 2 つの画像はいずれも彩度とコントラストを増してあります。
3 番目の図の B の画像はキャリブレートもプロファイルもしていないモニターで
<EM>見掛け上</EM> 自然な色合いにすべく <EM>緑</EM> が追加されました。
しかもこの画像の影に明るさが加えられ、 <EM>見掛け上</EM>
影のつぶれが緩和されたようになりました
(実際ははじめから影のつぶれなどありません)。
</P>
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-3A.jpg" style="float: left;">
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-3B.jpg" style="float: left;"></li>
<div class="clear"></div>
<p>When viewed on the respective monitors on which they were edited, #3A and #3B above will look very similar. But in fact they are very, very different. Image #3B actually has been edited to give it a green color cast, which makes it <em>look</em> neutral on a monitor that shows images with a "faux" magenta color cast.<br/>
双方のモニターで表示しながら画像をそれぞれ編集したあとのようすは上図 3 番目でご覧のとおり、
画像の A (色管理あり) と B (色管理なし) はかなりそっくり同じに見えます。
しかし本当は随分と異なっています。
3 番目の図の B の画像は緑がかった色合いになっており、
これはマゼンタ寄りに偏ったモニター表示で <EM>見掛け上</EM> 自然に見せようと加工された結果です。
</P>
<li><p>In a color-managed workflow, you can expect a print to more or less match what you see on your calibrated and profiled monitor (see #3A above and #4A below).<br/>
印刷物も色管理をとり入れた作業工程であれば、
キャリブレートとプロファイルが済んだモニターで見たままの仕上りが多少なりとも期待できます
(上図 3 番目の画像 A と下図 4 番目の画像 A をご覧ください)。
</P>
<p> If you don't use color management, then as long as your goal is to enjoy looking at an image displayed on the same uncalibrated monitor that you used to edit the image, all is well. But as soon as you send the image that was edited without using color management (#3B above) out to a color-managed professional print shop for printing (see #4B below), it becomes glaringly obvious that:<br/>
色管理をせずキャリブレートしていないモニターを使っていたとしても、
画像の加工と閲覧のいずれもそのモニター上で行なうだけでよいのであれば何も問題ありません。
しかし色管理をせず編集した画像 (下図 4 番目の画像 B) を、
色管理の行なわれている印刷業者へ印刷依頼に出したなら (下図 4 番目の画像 B)、
たちまち画像につぎの 2 つの問題が顕著に現れてきます。
</P>
<ul>
<li>To counteract the uncalibrated monitor's "faux" magenta color cast (see #2B above), green was added (see #3B above), making the image <em>look</em> neutral, when in reality the originally neutral image has been given a green color cast.<br/>
キャリブレートできておらずマゼンタに偏った色合いのモニターで表示され
(上図 2 番目の画像 B)、 これを直そうとして <EM>見掛け上</EM>
自然な色合いになるよう緑が追加された (上図 3 番目の画像 B) 結果、
元から自然な色合いだったはずの画像が緑がかったものになってしまっている。
</LI>
<li>Also, to counteract the uncalibrated monitor's "faux" crushed shadows (see #2B above), the shadows were lightened (see #3B above), producing shadows that in reality are lighter than intended.<br/>
しかも、 キャリブレートできておらず影がつぶれて見えるモニターで表示され
(上図 2 番目の画像 B)、 これを直そうとして影に明るさが追加された
(上図 3 番目の画像 B) 結果、
実際は影の表現が実感よりも明るくなってしまっている。
</LI>
</UL>
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-4A.jpg" style="float: left;">
<img src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/snow-tree-4B.jpg" style="float: left;"></li>
<div class="clear"></div>
</ol>
<p>To summarize, if you don't use proper color management when you edit images in your digital darkroom, then you don't know whether you are color correcting the actual image colors or merely compensating for idiosyncracies in how your uncalibrated and unprofiled monitor displays those colors.<br/>
まとめれば、
ご自分のディジタル暗室で適切に色管理をしないまま画像の編集を行なうと、
画像の本来の色を修正しているのかそれとも単にキャリブレートもプロファイリングもできていないモニターが表示する色の特異な癖の代償を払わされているだけなのかわかりませんよということです。
</P>
</figure>
<h3>Color management today: calibrate and then profile<br/>
いまどきの色管理はキャリブレートしてプロファイルする</H3>
<p>To recap:<br/>
まずはおさらい。
</P>
<Ul>
<li>Device calibration <em>alters</em> a device's color reproduction characteristics to meet target reproduction characteristics.<br/>
デバイスのキャリブレーションとはデバイスが色を再現する特性を <EM>補正</EM>
して特定の特性に合わせることです。
</LI>
<li>An ICC profile <em>describes</em> a device's color reproduction characteristics.<br/>
ICC プロファイルはデバイスが色を再現する特性を <EM>記述</EM> するしくみです。
</LI>
</UL>
<p><b>Even in an ICC profile color managed work flow, device calibration is a critically important part of overall color management.</b> For optimal color reproduction:<br/>
<B>
作業工程の色管理に ICC プロファイルを使用しても、
依然としてデバイスのキャリブレーションが色管理のあらゆる場面で決定的に重要な役割を担います。
</B>
よりよく色を再現させるために、
</P>
<ol>
<li>If a device can't be calibrated (eg cameras and scanners), profile it.<br/>
キャリブレートできないデバイス (例えばカメラやスキャナー) はプロファイリングします。
</LI>
<li>If a device <em>can</em> be calibrated (eg monitors and printers), first calibrate it and then profile it.<br/>
キャリブレートできるデバイス (例えばモニターやプリンター) はまずキャリブレートしてからプロファイリングします。
</LI>
</OL>
<h3>Using third party device ICC profiles<br/>
サードパーティー製のデバイス用 ICC プロファイルを利用
</H3>
<p>A "third party" device ICC profile is any device ICC profile that you didn't produce yourself. Using a third party profile is easier than making your own profile. However, if you use third party profiles, you should be aware of the following cautions and pitfalls:<br/>
サードパーティー製のデバイス用 ICC プロファイルとは、
自作以外のデバイス用 ICC プロファイルのことです。
流通しているプロファイルを入手し使用する方がご自身でプロファイリングするよりも手軽です。
しかしながら外部から入手したプロファイルを使用する場合、
つぎのような注意点と落とし穴に気を付けねばなりません。
</P>
<h4>Monitor profiles<br/>
モニターのプロファイル
</H4>
<p>Profiles for various monitors are readily available from the internet. However, there's no guarantee that a generic profile is a good match for your own monitor, simply because it's the result of profiling someone else's monitor. Putting issues of quality control aside, monitor manufacturers can and do change basic components in the middle of production runs. Also, monitor display characteristics drift over time. LCD monitors drift more slowly than CRT monitors, but all monitors need to be reprofiled on a schedule: weekly, monthly, or whenever you need to know the displayed color is as accurate as possible.<br/>
さまざまなモニター機種別プロファイルがネット上で配布されています。
しかしそのような普及版プロファイルが必ずしもお使いのモニターにぴったり適合する保証はどこにもありません。
理由は簡単で、 プロファイリングされたのは他人の所有するモニターだからです。
品質管理の問題を抜きにしても製造段階においてさえ、
メーカーがロットの途中から基本部品の変更を検討し実施した可能性もあります。
しかも、 ディスプレイモニターの特性は時間とともに推移します。
LCD モニターは CRT モニターよりもゆっくりと変質しますが、
やはりどのモニターも定期的にプロファイルを書き取り直す必要があります。
毎週でも毎月でも、
できれば表示される色が正確なのか気になるたびにいつでもプロファイリングするのです。
</P>
<h4>Printer and scanner profiles<br/>
プリンターとスキャナーのプロファイル
</H4>
<p>Printer and scanner profiles supplied by the manufacturer are generic and not necessarily a good match to your device. Nominally the same inks and paper can vary from batch to batch and also change with age and humidity. If you choose a different set of inks or different paper, you need a new profile. Scanner bulbs age and change color. If you are happy with the colors produced by your printer or scanner, great! But for the most accurate possible colors, you might want to profile your printers and scanners yourself.<br/>
プリンターとスキャナーのプロファイルは製造者が提供していますが一般的なものなので
必ずしもお使いのデバイスにぴったり適合するわけではありません。
紙とインクは名目上同一の製品であっても消費するかぎり常に変化がありますし時間の経過も湿度の影響も変化をもたらします。
一色でもインクの種類を変えたり、 別の種類の紙を使用することにしたのであれば、
プロファイルも交替が必要です。
スキャナーのバルブは古くなり色が変わります。
お使いのプリンターやスキャナーの出す色が満足のゆくものならすばらしいことです。
ただできうる限り最も正確な色が欲しいときにはやはりご自分でプリンターやスキャナーのプロファイリングをしたほうが良いのです。
</P>
<h4>Camera profiles<br/>
カメラのプロファイル
</H4>
<p>The dcraw camera matrices are used by a wide range of raw processors and are generally pretty good profiles. Compared to monitors, camera profiles are very shareable from one camera (of the same make and model) to the next, and also camera color reproduction characteristics are more or less stable over time. However, some people think the generic dcraw camera matrices produce somewhat anemic colors. <br/>
dcraw カメラマトリックスはロウ画像処理の分野で幅広く利用されておりおおむねかなり良好なプロファイルです。
モニター用と比べカメラのプロファイルは (同一メーカー同機種の)
カメラ同士の共有が非常にたやすく、
しかもカメラの色再現能力は時間の経過に対し特性がほぼ安定しています。
そうはいっても中には汎用 dcraw カメラマトリックスが出す色はいくぶん貧血ぎみなきらいがあると指摘する人もいます。
</P>
<p>If you shoot raw and you don't like the colors you get using the generic camera profile associated with your camera, you might want to try profiling your camera yourself. If you specialize in photographing art or commercial products where color accuracy is paramount, profiling your camera is essential.<br/>
ロウ撮影の画像が、
撮影したカメラ用の汎用プロファイルでは気に入らない色になる場合、
ご自分でプロファイリングをしてみてはいかがでしょうか。
色の正確さが何にも増して求められる写真芸術や商品生産を本業にしている方にとって、
使用するカメラのプロファイリングは欠かせません。
</P>
<A id="monitor"></A>
<h2>Calibrating and profiling your monitor<br/>
モニターのキャリブレーションとプロファイリングをしよう</H2>
<p>It's worth spending a bit of time talking about calibrating and profiling your monitor because when you use GIMP, you use your monitor to judge colors. <br/>
まだしばらくモニターのキャリブレーションとプロファイリングの話題におつきあいいただきますが、
モニター上で色を決めることになる GIMP のユーザーにとっては役立つ情報です。
</P>
<h3>Calibrating your monitor<br/>
自分用のモニターをキャリブレートする
</H3>
<p><a href="http://www.lagom.nl/lcd-test/">Calibration by visual comparison</a> is better than no calibration at all. But properly calibrating your monitor requires a dedicated measuring device, which can also be used to profile your monitor after you calibrate it.<br/>
<A href="http://www.lagom.nl/lcd-test/">実際の表示を見て比較するキャリブレーションの方法</A>
が他のどのキャリブレーションよりも優れています。
しかし適切にキャリブレートするためには特別な測定用デバイスを使用し、
モニターのキャリブレーションのあとのプロファイリングにもこれを使用する必要があります。
</P>
<p>Monitor calibration is done by measuring how your monitor currently displays colors, and then making appropriate alterations to bring it closer to some predefined target state. Alterations might include:<br/>
モニターのキャリブレーションは、
モニターが実際に表示している色を測りながら規定どおりの色になるまで適切な補正をしてゆき進めます。
補正にかかかわりそうな設定にはつぎのようなものがあります。
</P>
<ul>
<li>Setting the monitor brightness.<br/>
モニター輝度の設定
</LI>
<li>Setting a neutral axis that approximates a given tone reproduction curve, for example the sRGB tone reproduction curve, a given gamma curve, or the perceptually uniform l-star curve.<br/>
得られるカーブの近似をとる大まかな軸線の設定 &mdash;
たとえば sRGB 色調再現カーブのような色調再現曲線、 ガンマ曲線、 知覚的均等な L* 曲線のいずれかが対象
</LI>
<li>Setting a given white point, for example D50 or D65.<br/>
得られる白点の設定 &mdash; たとえば D50 や D65
</LI>
<li>Approximating a given color space gamut, for example the sRGB color gamut.<br/>
得られる色空間ガモットの近似操作 &mdash; たとえば sRGB 色空間のガモット
</LI>
</UL>
<p>There are two completely separate and independent ways to alter your monitor's color display characteristics:<br/>
モニターの色表示能力の特性を補正する方法は 2 通りあり、
両者に共通の手順は全くなく互いに独立しています。
</P>
<ol>
<li>You can <em>directly</em> affect the color display characteristics of your monitor by using your monitor controls to modify how how your monitor responds to the color intensity information sent to the monitor by your video card LUTs. Different monitors provide different levels of direct user control, ranging from no controls at all to very fine-grained controls.<br/>
モニターの色表示能力の特性を <EM>直接的</EM> に変更するにはモニターの制御盤を操作して、
そこでコンピューターのビデオカードの LUT から送られてくる色信号の強度と発色の相関を調整します。
モニター側の制御能力はピンからキリまであり、
一切変更ができない機種もあれば非常に細かな精度まで補正が効く機種までさまざまです。
</LI>
<li>You can <em>in</em>directly affect the color display characteristics of your monitor by modifying your video card LUTs to control the signals that are sent to the monitor. Modifications to the video card LUTs are stored as "vcgt" information, which typically is loaded into your video card upon starting your operating system.<br/>
モニターの色表示能力の特性を <EM>間接的</EM> に変更するにはビデオカードの LUT
を調整して、 それでモニターに送る信号を制御します。
ビデオカード LUT に加えた変更は「vcgt」と呼ばれる情報として保存され、
おもにオペレーティングシステムを起動したときにビデオカードから読み込まれます。
</LI>
</OL>
<p>What kinds of calibration should you do?<br/>
どちらのキャリブレーション方法を採ればよいのでしょう。
</P>
<ul>
<li>At one extreme, for a dedicated photography or digital art workstation the only LCD monitor calibration you might do is set brightness of the display, and you might not touch the video card LUTs at all.<br/>
極端な例ですが、 写真芸術やディジタルアート専用のワークステーションの
LCD モニターをキャリブレートするだけの場合、
ディスプレイの輝度設定を調整すればビデオカードの LUT は一切手をつけなくても良さそうです。
</LI>
<li>At the other extreme, for general usage and for the sake of your non-color-managed applications you might want to calibrate your monitor to produce display characteristics that match the sRGB color space as closely as possible. <br/>
また極端な例ですが、
反対に一般的なコンピューターの使い方をし色管理に関わりのないアプリケーションを使う用途ばかりであれば、
モニター側のキャリブレートを行ないそれでディスプレイの特性をできるかぎり sRGB
色空間に近づけるよう調整すれば良さそうです。
</LI>
</UL>
<p>Those of you who remember the old high-end CRTs know they were pretty accomodating of extensive calibration to whatever target state you had in mind. For LCDs, to avoid artifacts in displayed images, it's "<a href="http://argyllcms.com/doc/monitorcontrols.html">generally best to set all the LCD controls except back lighting brightness to their default settings. . . . [and] set the transfer characteristic and neutral axis, and leave the white point native</a>".<br/>
(未訳)
</P>
<h3>Calibrating and profiling your monitor with ArgyllCMS<br/>
ArgyllCMS によるキャリブレーションとプロファイリング
</H3>
<p><a href="http://argyllcms.com/">ArgyllCMS is an open source color management application</a> for calibrating and profiling monitors, and also for profiling scanners, cameras, and printers. ArgyllCMS is a command-line only program and is available for the Linux, Windows, and Apple operating systems. For the command-line averse, there are various GUI programs that make using ArgyllCMS a little less scary.<br/>
<A href="http://argyllcms.com/">オープンソースの色管理アプリケーション ArgyllCMS</A> は、
モニターのキャリブレーションとプロファイリングに使用できるほかスキャナーやカメラやプリンターのプロファイリングにも使用できます。
ArgyllCMS はコマンドライン入力方式のみのプログラムで、 Linux、 Windows、 Apple
などのオペレーティングシステムで動作します。
ArgyllCMS を GUI で操作できるプログラムが何種類か出回っており、
コマンドライン嫌いの人たちの恐怖をやわらげてくれます。
</P>
<p>For easy monitor calibration and profiling, ArgyllCMS provides a simple command for <a title="Adjusting, calibrating and profiling in one step." href="http://argyllcms.com/doc/Scenarios.html#PM1c">calibrating your monitor, loading any required vcgt information into your video card LUTs, and then profiling your display</a>. For advanced users, ArgyllCMS provides fine-grained multistep monitor calibration and profiling options. If you want to use ArgyllCMS to calibrate and profile your monitor, check the list of <a href="http://argyllcms.com/doc/ArgyllDoc.html">supported measuring devices</a> before purchasing a measuring device. <br/>
ArgyllCMS は手軽にモニターのキャリブレーションとプロファイリングができるよう簡単なコマンドを提供しており、
<A title="一挙に調整・キャリブレート・プロファイリング"
href="http://argyllcms.com/doc/Scenarios.html#PM1c">
これ一発でモニターがキャリブレートされ、
必要な vcgt 情報がビデオカード LUT に読み込まれ、
つづけてディスプレイのプロファイリングまでまとめて実行できます。
</A>
ArgyllCMS は上級者向けに、
モニターのキャリブレーションとプロファイリングを細かな精度で順を追って実行できるオプションも用意しています。
ArgyllCMS を使用してモニターのキャリブレートとプロファイリングをするために専用の測定デバイスを買い求める予定がある方は、
事前に <A href="http://argyllcms.com/doc/ArgyllDoc.html">
サポートされている測定デバイスの一覧</A>
をチェックしておきましょう。
</P>
<h3>Installing your monitor profile<br/>
モニターのプロファイルのインストール方法
</H3>
<p>Installing a monitor profile is actually a two-step process, <a title="Installing a display profile" href="http://argyllcms.com/doc/Scenarios.html#PM5">procedures for which vary from operating system to operating system</a>: first any required vcgt <em>calibration</em> information is loaded into the video card LUTs, and then the monitor profile is associated with the display. For convenience, the vcgt information is usually stored right inside the monitor profile as a special <dfn>vcgt tag</dfn>, but the vcgt tag is (and should be) completely ignored by color-managed applications like GIMP. <br/>
モニターのプロファイルのインストールする方法は実質的に 2 段階の工程からなり、
<A title="ディスプレイプロファイルのインストール方法"
href="http://argyllcms.com/doc/Scenarios.html#PM5">
その手順はオペレーティングシステムごとに違いがあります。
</A>
まず必須の <EM>キャリブレーション</EM> vcgt 情報をビデオカードの
LUT に読み込ませ、
そのつぎにモニターのプロファイルをディスプレイと関連づけます。
通常 vcgt 情報は便宜的にモニターのプロファイルのなかに特別な <DFN>vcgt タグ</DFN>
となって保存されますが、
GIMP をはじめ色管理を行なうアプリケーションは vcgt タグを全く無視しますしそうするはずです。
</P>
<p>For Linux users, in addition to ArgyllCMS, other tools that might be available for managing your display profile(s) depend on which software packages you've installed and which version of which desktop or window manager you might be using.<br/>
Linux ユーザー向けには ArgyllCMS の他にもディスプレイのプロファイルを管理するツールがあり、
どれが使えるかはインストール済みソフトウェアパッケージやお使いのウィンドウ/デスクトップマネージャーの種類によって異なります。
</P>
<h3>Telling GIMP which monitor profile to use<br/>
使用するモニタープロファイルは何か GIMP に認識させる
</H3>
<p>In GIMP's Color Management Preferences, you can tell GIMP to use the system-installed monitor profile. GIMP also lets you choose a monitor profile from disk for those times when you don't want to use the system-installed monitor profile. Just make sure that whatever monitor profile you tell GIMP to use was made using whatever vcgt information is loaded into your video card LUTs, or else you will see inaccurate colors on your screen.<br/>
GIMP の設定ダイアログのカラーマネジメントのページで、
システムにインストールされているモニタープロファイルを優先して使用するよう GIMP
に指示できます。
またここではシステムにインストールしていないモニタープロファイルをディスクから選び
GIMP に使用させることもできます。
ただしたとえどんなモニタープロファイルを GIMP で使用するにしても、
何であれお使いのビデオカードの LUT に読み込まれている vcgt
情報に基づいて作成されたプロファイルでなければ画面に現れる色が不正確なものとなります。
</P>
<a id="color"></a><h2>What is color?<br/>
色とは何か</H2>
<h3>Light and color<br/>
光と色</H3>
<p>Light is the miracle that makes color possible, but color isn't out there in the world in the same tangible way that light is. Rather different wavelengths and intensities of light are processed by light receptors in the eyes (cones and rods), and the resulting signals are sent via the optic nerves to the brain for further processing and interpretation. The three different types of cones in the eye have overlapping sensitivities to different wavelengths of light, which is the basis for our perception of color.<br/>
光は色を色たらしむる奇跡でありますが、
色は光のような自明の理をもってこの世に存在しているのではありません。
ある程度の差異がある光の波長や振幅を眼の光受容器 (網膜と桿体) が処理し、
生成された信号が視神経を通じて脳へ送られさらなる処理と解釈が施されます。
眼球内の 3 種類の桿体が部分的に重なる分担範囲で光の波長の違いを認識し、
これが我々人類の色覚の基礎をなします。
</P>
<h3>Describing color perception with mathematics: CIEXYZ and CIELAB<br/>
色覚の数理的描写方法 &mdash; CIEXYX と CIELAB
</H3>
<p>In the late 1920s William David Wright and John Guild independently conducted a series of color matching experiments that mapped out all the colors the average human can see. In 1931 color scientists used the results of the Wright and Guild experiments to create the 1931 CIEXYZ color space ("XYZ" for short).<br/>
1920 年代末ごろ William David Wright 氏と John Guild 氏がそれぞれ別々に、
平均的なヒトが視認できるすべての色を配置する一連の実験を行いました。
1931 年に色彩学者たちは Wright と Guild の実験の成果をもとに「1931 CIE XYZ 色空間」
(略して XYZ) を作成します。
</P>
<p>The XYZ color space is based on the physical properties of light. Light combines in a linear fashion, but our eyes and brain perceive color in a very nonlinear fashion. We are much more sensitive to changes in the ambient light level when the light level is relatively low, and less sensitive as the amount of light increases. Similar considerations apply in our perception of more and less saturated colors, and of changes in hue (eg red to orange to yellow). So various mathematical derivations of the XYZ rcolor space have been proposed in an attempt to create a <dfn>perceptually uniform</dfn> color space. Currently the most widely used perceptually uniform reference color space is the 1976 CIELAB color space ("LAB" for short). <br/>
XYZ 色空間は光の物理的属性に基づいています。
光の合成は線形的に行われますが、 視覚や脳は色をかなり非線形的に知覚します。
我々の周囲の光度の変化に対する視力は、
比較的光度が低ければより敏感になり光度が高くなるにつれて鈍感になります。
同様な傾向が多少彩度のある色や色相の変化 (例えば赤→橙→黄)
に対する知覚においても見られます。
そのため XYZ 色空間からさまざまな数理展開が提案され <DFN>知覚的均等</DFN>
な色空間の作成が試みられます。
現在最もひろく利用されている知覚的均等標準色空間は「1976 CIELAB 色空間」
(略して LAB) です。
</P>
<h3>Reference color spaces<br/>
標準色空間
</h3>
<p>Every color that you can see out there in the real world, whether it's the color of a flower in your hand or the color of a flower displayed on your computer screen, has a single location inside the XYZ color space. Because LAB is a mathematical transform of XYZ, the same is true of the LAB color space. Which means that if you specify a color by giving its XYZ or LAB color space coordinates, then everyone else knows exactly what color you mean. <br/>
この世界中の目で見えるどんな色も、
それが手に持っている花の色であれコンピューター画面に映し出された花の色であれ、
XYZ 色空間のなかではひとつひとつの座標となります。
LAB は XYZ の数理変換によるものなので、 この事実は LAB 色空間でも同じです。
したがって XYZ または LAB 色空間から一点の座標でもって色を示せば、
まったく同じ色を他の人に伝えられます。
</P>
<p>Because all colors can be precisely located in the XYZ and LAB color spaces, we call these color spaces <dfn>reference color spaces</dfn>. There are other reference color spaces, all of which are mathematically derived from XYZ. For example, the xyY color space completely separates luminance ("Y") from chromaticity ("xy"). <br/>
すべての色が厳密に XYZ または LAB 色空間上で配置できることから、
このような色空間は <DFN>標準色空間</DFN> と呼ばれています。
他にも標準色空間の種類があり、 いずれも XYZ の数理展開です。
例えば xyY 色空間は輝度「Y」と色度「xy」を完全に分離します。
</P>
<p>If you know a color's location in any one of the reference color spaces, you can calculate its location in any of the other reference color spaces.<br/>
色をいずれかの種類の標準色空間上の座標で示せれば、
数値計算によって他のどんな種類の標準色空間でも座標が算出できます。
</P>
<h3>Different kinds of color spaces<br>
色空間の種類
</H3>
<li>RGB</li>
<li>CMYK</li>
<li>Gray</li>
<li>n-space</li>
<a id="color-gamuts"></a><h2>Color gamuts<br/>
色ガモット</h2>
<h3>Device color gamuts<br/>
デバイスの色ガモット
</H3>
<p>If every color reproduction device could reproduce all the colors that the average human can see ("all the real colors", for short), color management would be a lot simpler. The cold hard reality is that devices vary greatly in the colors they can reproduce. <br/>
色再現・発色を行なうあらゆるデバイスがいずれも平均的なヒトの目で見えるすべての色
(要するにすべての天然色) を再現可能だとしたら、
色管理はとてつもなく簡単になるはずです。
現実は冷酷で、 再現できる色により各デバイスに大きな違いがあります。
</P>
<p>The sum total of all the colors a device can reproduce is called its <dfn>color gamut</dfn>. Color gamuts are actually three-dimensional <em>volumes</em> of various sizes and shapes as located inside one of the reference color spaces.<br/>
デバイスごとにそれが再現できるすべての色の範囲を <DFN>色ガモット</DFN>
といいます。
実際の色ガモットは三次元の <EM>立体</EM> 図形をなしており、
いずれかの種類の標準色空間のなかでさまざまな大きさや形をもって表されます。
</P>
<p>Figure 11.2 shows sample color gamuts for selected devices inside the 1976 CIELAB color space. The white wire frames show various device color gamuts. The multicolored blobs are the sRGB color space, shown for comparison. For the color gamuts on the left (the three display profiles), the view is looking straight down the CIELAB L* axis. The color gamuts on the right (one camera and two printer profiles) have been rotated in various directions in order to emphasize that color gamuts are three-dimensional:<br/>
図11.2 は試しにいくつかのデバイスの色ガモットを
1976 CIELAB 色空間内に示したものです。
白のワイヤーフレームがデバイスの色ガモットを表しています。
比較用に sRGB 色空間を多色図形で重ねて示しました。
左に並べた色ガモット (3 台のディスプレイのプロファイル) は CIELAB
の L* 軸に沿ってまっすぐ見下ろす視点で描いてあります。
右に並べた色ガモット (カメラ 1 台とプリンター 2 機のプロファイル)
はそれぞれ色ガモットが三次元立体であることを強調するためいずれかの方向に回転して描いてあります。
</P>
<figure>
<figcaption>Figure 11.2. Sample Device Color Gamuts (white wire frames) compared to sRGB (multicolored blobs)<br/>
図 11.2 デバイスの色ガモット標本 (白ワイヤーフレーム) と sRGB (多色図形) の比較
<ol>
<li>Laptop display profile<br/>
ラップトップディスプレイのプロファイル
</LI>
<li>Consumer grade LCD profile<br/>
一般消費者向け LCD のプロファイル
</LI>
<li>Professional photo-editing LCD profile<br/>
専門家向け写真編集用 LCD のプロファイル
</LI>
<li>Digital camera profile<br/>
ディジタルカメラのプロファイル
</LI>
<li>Printer profile (prints for matting and framing)<br/>
プリンターのプロファイル (艶消し紙と額縁写真)
</LI>
<li>Printer profile (newspaper stock)<br/>
プリンターのプロファイル (新聞の切り抜き)
</LI>
</OL>
</figcaption>
<img
src="http://ninedegreesbelow.com/gimpgit/gimpdoc/pictures/devices.jpg">
</figure>
<p>As Figure 11.2 shows, device color gamuts vary considerably. Three points are worth emphasizing:<br/>
図 11.2 でご覧のとおりデバイスの色ガモットは目に見えてはっきり違いがあります。
これについてとくにつぎの 3 つのことが言えます。
</P>
<ol>
<li><b>Coverage of the sRGB color space varies considerably from one display device to another</b>. In Figure 11.2:<br/>
<B>
sRGB 色空間に占める有効範囲はデバイスごとにはっきりと違いがあります。
</B>
図 11.2 においてもつぎのことが確認できます。
<UL>
<li>The laptop display (#1) holds 59% of the sRGB color space.<br/>
ラップトップのディスプレイ (1 番) は sRGB 色空間の 59% を占めます。
</LI>
<li>The consumer grade LCD monitor (#2) holds 88% of the sRGB color space.<br/>
一般消費者向け LCD モニター (2 番) は sRGB 色空間の 88% を占めます。
</LI>
<li>The professional photo-editing LCD (#3) holds 100% of the sRGB color space.<br/>
専門家向け写真編集用 LCD (3 番) は sRGB 色空間の 100% を占めます。
</LI>
</UL>
</li>
<li><b>Devices vary greately in what percentage of all real colors are included in their profile color gamuts.</b> In Figure 11.2:<br/>
<B>
デバイスのプロファイルの色ガモットがすべての天然色に占める割合はデバイスによって大きく異なります。
</B>
図 11.2 においてもつぎのことが確認できます。
<UL>
<li>The sRGB color gamut itself is not very large, holding only 35% of all real colors.<br/>
sRGB 色ガモット自体があまり大きくなく、
総天然色に占める割合は 35% しかありません。
</LI>
<li>Even the camera profile color gamut (#4), which completely engulfs the sRGB color space, doesn't hold all real colors. <br/>
カメラのプロファイルの色ガモット (4 番) でさえ、
sRGB 色空間を丸ごと飲み込んでいるにもかかわらず天然色のすべてを占めてはおりません。
</LI>
<li>The color gamut for the newspaper stock printer profile (#6) is extremely small, holding 16% of the sRGB color space and a miniscule 6% of all real colors.<br/>
プリンターの新聞切り抜き用プロファイル (6 番) の色ガモットは極端に小さく、
sRGB 色空間の 16%、 総天然色のたったの 6% を占めます。
</LI>
</UL>
</li>
<li><b>Device color gamuts vary in shape as well as size.</b> For example, in Figure 11.2 the profile for the printer used to make prints for matting and framing (#5) is almost as large as the sRGB color space, but has a completely different shape:<br/>
<B>
デバイスの色ガモットは形状も大きさも違いがあります。
</B>
例えば、 図 11.2 においてプリンターの艶消し額縁写真用プロファイル (5 番)
は sRGB 色空間とほぼ同じ大きさですが形がまったく異なります。
<UL>
<li>On the one hand, roughly 30% of the sRGB color gamut (the more saturated reds, magentas, and blues) can't be printed on this printer.<br/>
一方ではざっと 30% の sRGB 色ガモットが (彩度高めの赤・マゼンタ・青の範囲で)
このプリンターは印刷できません。
</LI>
<li>On the other hand, roughly 20% of the color gamut that this printer <em>can</em> print (the more saturated yellows, greens, and teals) falls completely outside the sRGB color gamut.<br/>
他方でこのプリンターの色ガモットのだいたい 20 % は
(彩度高めのイエロー・緑・深緑青の範囲で) まったく sRGB 色ガモットにない色を
<EM>発色可能</EM> です。
</LI>
</UL>
</li>
</ol>
<h3>RGB working space color gamuts<br/>
RGB 作業色空間のガモット
</H3>
<li>The historical link between standard RGB working spaces and monitor profiles</li>
<li>sRGB vs ProPhotoRGB vs real world colors captured by digital cameras</li>
<a id="proofing"></a><h2>Soft proofing<br/>
ソフトプルーフ</H2>
<p>The art of dealing with mismatched color gamuts when sending an image from one device or working space to another is called <dfn>soft proofing</dfn>.<br/>
別のデバイスや作業色空間に画像を送ったときに生じる色ガモットの食い違いに対処する技術を
<DFN>ソフトプルーフ</DFN> と呼びます。
</P>
<a id="gimp-specific"></a><h2>GIMP specific<br/>
GIMP のみに関わる話題</h2>
<h3>Open/Assign/Convert</h3>
<li>what to do about images with and without embedded ICC profiles</li>
<li>assigning ICC profiles and converting from one profile to another</li>
<h3>Unbounded mode high bit depth</h3>
<li>high bit depth unbounded mode ICC profile conversions</li>
<li>unbounded mode image editing: levels vs curves, gamma vs linear</li>
<li>How high bit depth GIMP gets around the problem of clipping when converting to sRGB</li>
<h3>GIMP color management options</h3>
<li>global and per image color management options</li>
<a id="more-information"></a><h2>Links to more information</h2>
<!--
<h3>Types of ICC profiles</h3>
<li>lut vs matrix</li>
-->
</div><!-- gimppad -->
</body>
</html>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment