More experiments with sRGB gamut boundary in L * a * b * space

作者史蒂夫·埃丁,2022年5月3日

Look at 68 times (the past 30 days) |0个赞|Three comments

            我仍然和RGB色域计算
            ^ * ^ ^ * b * $L $
            Space. (See my last post on this topic.
            "Visualizing the out - of - gamut colors in a Lab curve."
            ) Today 's post the features some new gamut - related visualizations, plus some computational tricks involving gamut boundaries and rays in
            ^ * ^ ^ * b * $L $
            空间。
           

            首先,这里是另一种交流的想法in-gamut区域
            $(^ ^ *, b *) $
            飞机随
            L ^ * $美元
            (perceptual lightness). For 9 values of
            L ^ * $美元
            (10年,20年,……,90), I'll compute a 2-D
            $(^ ^ *, b *) $
            Gamut mask by brute - forcing it. Then, I "l l use overlaid contour plots to show the variation in gamut boundaries.
           

              A = - 110:. 1:11 0;
             

              B = - 110:. 1:11 0;
             

              L = 10:20:90;
             

              (aa、bb、LL) = meshgrid (a、b L);
             

              持有on
             

              为k = 1:长度(L)
             

              rgb = lab2rgb(猫(LL (:,:, k), aa (:,:, k), bb (:,:, k)));
             

              Mask = all (< = RGB (0) & (RGB < = 1), 3) * 2-1 + L (k);
             

              轮廓(a, b,面具,[L (k) L (k)], LineColor = [。8。8。8),线宽= 1.5,ShowText = true,...
             

              LabelSpacing = 288)
             

              结束
             

              持有从
             

              The axisequal
             

              The gridon
             

              盒子on
             

              Xlabel ("A *")
             

              Ylabel (“b *”)
             

              The title (“色域边界(a, b)飞机数的值L *”)
             

            这是另一个可视化的概念。人们经常显示颜色
            $(^ ^ *, b *) $
            Plane, to difference an idea of the fancy of
            $ ^ * $
            The and
            ^ * $ $ b
            , but that doesn 't communicate very well, the idea that there are usually multiple colors, corresponding to various
            L ^ * $美元
            Values, at any one
            $(^ ^ *, b *) $
            The location. Below, I show both the brightest - gamut color and in the darkest in - gamut color at each
            $(^ ^ *, b *) $
            The location.
           

              一个= 110:110;
             

              b = 110:110;
             

              Aa, bb] [= meshgrid (a, b);
             

              L_max = 0(大小(aa));
             

              L_min = 0(大小(aa));
             

              为P = 1: size (aa, 1)
             

              为Q = 1: size (bb, 1)
             

              [L_min (p, q), L_max (p, q)] = Lrange (aa (p, q), bb (p, q));
             

              结束
             

              结束
             

              rgb = lab2rgb(猫(3 L_max aa、bb));
             

              figure
             

              Tiledlayout (1, 2)
             

              nexttile
             

              imshow (rgb, XData =([1]结束),YData = b(结束[1]))
             

              The axisThe xy
             

              The axison
             

              xlabel一个
             

              ylabelb
             

              The title (“聪明的in-gamut颜色”)
             

              rgb_min = lab2rgb(猫(3 L_min aa、bb));
             

              nexttile
             

              Imshow (rgb_min, XData = a (end [1]), YData = b (end [1]))
             

              The axison
             

              The axisThe xy
             

              xlabel一个
             

              ylabelb
             

              The title ("Darkest in - gamut color")
             

            Next, I find myself sometimes wanting to draw a ray in
            ^ * ^ ^ * b * $L $
            空间和发现色域边界的位置沿射线。为此,我写了一个简单的效用函数(
            findNonzeroBoundary
            下图)执行一个二叉搜索找到一个函数从积极到0。然后,我写了一些匿名函数找到所需的色域边界点。具体地说,我很感兴趣这个问题:对于一个给定的
            L ^ * $美元
            The value and given a
            $(^ ^ *, b *) $
            平面的角度,
            h
            ,什么是in-gamut颜色最大浓度,
            c
            , or short the from
            $$(0, 0)
            In the
            $(^ ^ *, b *) $
            飞机吗?
           

            Fair warning
            : the code below gets tricky with anonymous functions provides. You took hate it. If so, I totally understand, and I hope You 'l l forgive me. : -)
           

            我将首先创建一个匿名函数,转换
            H $$L ^ * c
            The to
            ^ * ^ ^ * b * $L $
            :
           

              lch2lab = @(禄)[禄(1)华尔街日报(2)* cosd(华尔街日报(3)华尔街日报(2)*信德(华尔街日报(3)));
             

            Next, here is an anonymous function that returns been or not a particular
            ^ * ^ ^ * b * $L $
            The point is in gamut.
           

              InGamut = @ (lab) all (0 < = lab2rgb (lab), 2) & all (lab2rgb (lab) < = 1, 2);
             

            最后,第三个匿名函数使用
            findNonzeroBoundary
            找到我感兴趣的色域边界点。
           

              MaxChromaAtLh = @ (L, h) findNonzeroBoundary (@ (c) inGamut (lch2lab ([L] c h)), 0, 200);
             

            让我们锻炼这个函数找到高色深色
            $h = 0 ^ {\ circ} $
            。
           

              L = 35;
             

              H = 0;
             

              c = maxChromaAtLh (L、h)
             

               C = 61.0229

            这是那个颜色是什么样子的。
           

              rgb_out = lab2rgb (lch2lab ([L c h]));
             

                  Rgb_out =
                  1 x 3
                 

                  0.6414 0.0001 0.3316

              figure
             

              ColorSwatches (rgb_out)
             

              The axisequal
             

              The axis从
             

            (函数
            colorSwatches
            是来自
            Digital Image Processing Using MATLAB
            , and it is included in the
            MATLAB代码文件的书
            。)
           

            What happens when we try to find a high chroma color, at the same hue Angle, that home is bright dark?
           

              L = 90;
             

              H = 0;
             

              c = maxChromaAtLh (L、h)
             

               C = 14.8666

            You can can see that the maximum
            c
            价值低得多的价值就越高
            L ^ * $美元
            . What does it look like?
           

              rgb_out = lab2rgb (lch2lab ([L c h]));
             

              figure
             

              ColorSwatches (rgb_out)
             

              The axisequal
             

              The axis从
             

            当我在做这些实验准备这篇博客,我的初衷是给几个不同值的例子
            h
            The and
            L ^ * $美元
            。但我不能停止!这太有趣,我总是在不同的值。
           

            After about 15 minutes or so, whenever I decided it best to write some simple loops to generate a relatively large number of
            (L ^ *, h)美元
            组合。这是代码生成高
            c
            Colors for a variety of
            (L ^ *, h)美元
            Combinations.
           

              dh = 30;
             

              H = - 180: dh: 150;
             

              L = didst: 95;
             

              dL = 15;
             

              RGB = zeros (length (h), length (L), 3);
             

              为q = 1:长度(L)
             

              为K = 1: length (h)
             

              C = maxChromaAtLh (L (q), h (k));
             

              rgb (k问:)= lab2rgb (lch2lab ([L (q) c h (k))));
             

              结束
             

              结束
             

            这里是查看所有这些颜色的代码在一个网格,标记
            h
            The and
            L ^ * $美元
            轴。
           

              rgb2 =重塑(fliplr (rgb), [], 3);
             

              P = colorSwatches (rgb2, [length (L) length (h)]);
             

              p。XData = (p。XData - 0.5) * (dh / 1.5) + h (1);
             

              p。YData = (p。YData - 0.5) * (dL / 1.5) + L (1);
             

              Xticks (h);
             

              Xlabel (“h”)
             

              Yticks (L)
             

              Ylabel ("L *")
             

              The gridon
             

              The title (Most "Highest chroma (saturated) colors for marketers and L * h values")
             

The Utility Functions provides

              函数[L_min, L_max] = Lrange (a, b)
             

              The arguments
             

              一个{mustBeFloat} (1, 1)
             

              b{mustBeFloat} (1, 1)
             

              结束
             

              L = 0-0. 01:10 0;
             

              A = a * ones (size (L));
             

              B = b * ones (size (L));
             

              实验室= [L;一个;b];
             

              rgb = lab2rgb(实验室);
             

              gamut_mask = ((0 < = rgb) & (rgb < = 1), 2);
             

              j =找到(gamut_mask 1“第一”);
             

              k =找到(gamut_mask 1“最后一次”);
             

              如果isempty (j)
             

              L_min = NaN;
             

              L_max = NaN;
             

              其他的
             

              L_min = L (j);
             

              L_max = L (k);
             

              结束
             

              结束
             

              函数x = findNonzeroBoundary (f, x1, x2, abstol)
             

              The arguments
             

              f(1, 1) function_handle
             

              The x1{mustBeFloat} (1, 1)
             

              x2{mustBeFloat} (1, 1)
             

              abstol{mustBeFloat} (1, 1)= 1的军医
             

              结束
             

              如果(f (x1) = = 0) | | (f (x2) ~ = 0)
             

              Error (”功能必须在初始非零起点和零初始终点。”)
             

              结束
             

              Xm = mean ([x1 x2]);
             

              如果abs (xm - x1) / max (abs (xm), abs (x1) < = abstol
             

              x = x1;
             

              elseif(f (xm) = = 0)
             

              x = findNonzeroBoundary (f, x1, xm);
             

              其他的
             

              X = findNonzeroBoundary (f, xm, x2);
             

              结束
             

              结束
             

要发表评论,请点击hereLog in to your MathWorks account or create a new account.

史蒂夫与MATLAB图像处理
图像处理的概念、算法和MATLAB

图像处理的概念、算法和MATLAB

More experiments with sRGB gamut boundary in L * a * b * space

The Utility Functions provides

评论

The Utility Functions provides

另请参阅

评论

选择一个网站

美洲

欧洲

Asia Pacific