% CMAP Library of perceptually uniform colour maps % % Usage: 1: [map, name, desc] = cmap(I, param_name, value ...) % 2: cmap % 3: cmap(str) % % Arguments for Usage 1: % % I - A string label indicating the colour map to be generated or a % string specifying a colour map name or attribute to search % for. Type 'cmap' with no arguments to get a full list of % possible colour maps and their corresponding labels. % % labels: 'L1' - 'L19' for linear maps % 'D1' - 'D13' for diverging maps % 'C1' - 'C11' for cyclic maps % 'R1' - 'R4' for rainbow maps % 'I1' - 'I3' for isoluminant maps % % labels for generating maps for the colour blind: % 'CBL1' - 'CBL4' Linear maps for protanopic and deuteranopic viewers. % 'CBD1' - 'CBD2' Diverging maps for protanopic and deuteranopic viewers. % 'CBC1' - 'CBC2' Cyclic maps for protanopic and deuteranopic viewers. % 'CBTL1' - 'CBTL3' Linear maps for tritanopic viewers. % 'CBTD1' Diverging map for tritanopic viewers. % 'CBTC1' - 'CBTC2' Cyclic maps for tritanopic viewers. % % Some colour maps have alternate labels for convenience and readability. % In many cases the hue string (see below) acts as an alternate label. % >> map = cmap('L1') or map = cmap('grey') will produce a linear grey map. % >> map = cmap('D1') or map = cmap('BWR') blue-white-red diverging map. % >> cmap % With no arguments lists all colour maps and labels. % % Possible param_name - value options: % % 'chromaK' - The scaling to apply to the chroma values of the colour map, % 0 - 1. The default is 1 giving a fully saturated colour map % as designed. However, depending on your application you may % want a colour map with reduced chroma/saturation values. % You can use values greater than 1 however gamut clipping is % likely to occur giving rise to artifacts in the colour map. % 'N' - Number of values in the colour map. Defaults to 256. % 'shift' - Fraction of the colour map length N that the colour map is % to be cyclically rotated, may be negative. (Should only be % applied to cyclic colour maps!). Defaults to 0. % 'reverse' - If set to 1 reverses the colour map. Defaults to 0. % 'diagnostics' - If set to 1 displays various diagnostic plots. Note the % diagnostic plots will be for the map _before_ any cyclic % shifting or reversing is applied. Defaults to 0. % % The parameter name strings can be abbreviated to their first letter. % % Returns: % map - Nx3 rgb colour map % name - A string giving a nominal name for the colour map % desc - A string giving a brief description of the colour map % % % Usage 2 and 3: cmap(str) % % Given the large number of colour maps that this function can create this usage % option provides some help by listing the numbers of all the colour maps with % names containing the string 'str'. Typically this is used to search for % colour maps having a specified attribute: 'linear', 'diverging', 'rainbow', % 'cyclic', or 'isoluminant' etc. If 'str' is omitted all colour maps are % listed. % % >> cmap % lists all colour maps % >> cmap('diverging') % lists all diverging colour maps % % Note the listing of colour maps can be a bit slow because each colour map has to % be created in order to determine its full name. % % % Colour Map naming convention: % % linear_kryw_5-100_c67_n256 % / / | \ \ % Colour Map attribute(s) / | \ Number of colour map entries % / | \ % String indicating nominal | Mean chroma of colour map % hue sequence. | % Range of lightness values % % In addition, the name of the colour map may have cyclic shift information % appended to it, it may also have a flag indicating it is reversed. % % cyclic_wrwbw_90-40_c42_n256_s25_r % / \ % / Indicates that the map is reversed. % / % Percentage of colour map length % that the map has been rotated by. % % * Attributes may be: linear, diverging, cyclic, rainbow, or isoluminant. A % colour map may have more than one attribute. For example, diverging-linear or % cyclic-isoluminant. % % * Lightness values can range from 0 to 100. For linear colour maps the two % lightness values indicate the first and last lightness values in the % map. For diverging colour maps the second value indicates the lightness value % of the centre point of the colour map (unless it is a diverging-linear % colour map). For cyclic and rainbow colour maps the two values indicate the % minimum and maximum lightness values. Isoluminant colour maps have only % one lightness value. % % * The string of characters indicating the nominal hue sequence uses the following code % r - red g - green b - blue % c - cyan m - magenta y - yellow % o - orange v - violet % k - black w - white j - grey % % ('j' rhymes with grey). Thus a 'heat' style colour map would be indicated by % the string 'kryw'. If the colour map is predominantly one colour then the % full name of that colour may be used. Note these codes are mainly used to % indicate the hues of the colour map independent of the lightness/darkness and % saturation of the colours. % % * Mean chroma/saturation is an indication of vividness of the colour map. A % value of 0 corresponds to a greyscale. A value of 50 or more will indicate a % vivid colour map. % % For more information about the design of these colour maps see: % Peter Kovesi. Good Colour Maps: How to Design Them. % https://arxiv.org/abs/1509.03700 % See also: % https://peterkovesi.com/projects/colourmaps % % See also: COLORCET, EQUALISECOLOURMAP, VIEWLABSPACE, SINERAMP, CIRCLESINERAMP, COLOURMAPPATH % Adding your own colour maps is straightforward. % % 1) Colour maps are almost invariably defined via a spline path through CIELAB % colourspace. Use VIEWLABSPACE to work out the positions of the spline % control points in CIELAB space to achieve the colour map path you desire. % These are stored in an array 'colpts' with columns corresponding to L a and % b. If desired the path can be specified in terms of RGB space by setting % 'colourspace' to 'RGB'. See the ternary colour maps as an example. Note % the case expression for the colour map label must be upper case. % % 2) Set 'splineorder' to 2 for a linear spline segments. Use 3 for a quadratic % b-spline. % % 3) If the colour map path has lightness gradient reversals set 'sigma' to a % value of around 5 to 7 to smooth the gradient reversal. % % 4) If the colour map is of very low lightness contrast, or isoluminant, set % the lightness, a and b colour difference weight vector W to [1 1 1]. % See EQUALISECOLOURMAP for more details % % 5) Set the attribute and hue sequence strings ('attributeStr' and 'hueStr') % appropriately so that a colour map name can be generated. Note that if you % are constructing a cyclic colour map it is important that 'attributeStr' % contains the word 'cyclic'. This ensures that a periodic b-spline is used % and also ensures that smoothing is applied in a cyclic manner. Setting the % description string is optional. % % 6) Run CMAP specifying the number of your new colour map with the diagnostics % flag set to one. Various plots are generated allowing you to check the % perceptual uniformity, colour map path, and any gamut clipping of your % colour map. % % Reference: % Peter Kovesi. Good Colour Maps: How to Design Them. % arXiv:1509.03700 [cs.GR] 2015 % https://arxiv.org/abs/1509.03700 % Copyright (c) 2013-2020 Peter Kovesi % Centre for Exploration Targeting % The University of Western Australia % % peterkovesi.com % % Permission is hereby granted, free of charge, to any person obtaining a copy % of this software and associated documentation files (the "Software"), to deal % in the Software without restriction, subject to the following conditions: % % The above copyright notice and this permission notice shall be included in % all copies or substantial portions of the Software. % % The Software is provided "as is", without warranty of any kind. % December 2013 Original version % March 2014 Various enhancements % August 2014 Catalogue listing % September 2014 Provision for specifying paths in RGB, cyclic shifting and % reversing % October 2014 Renamed to CMAP (from LABMAPLIB) and all the failed % experimental maps cleaned out and retained maps renumbered % in a more coherent way % June 2016 Minor adjustments to some maps to minimise gamut clipping, % this was causing some artifacts % March 2017 Minor adjustement to L3 and L4 heatmaps. Addition of % alternate names for some maps and ensured that there were % description strings for all maps. % August 2017 Addition of colour maps for colour blind % viewers. Protanopic, deuteranopic and tritanopic. % October 2017 Slight update to 'L9' to make the blue section less % dominant. % April 2018 Added L16 black-blue-green-yellow and L17, L18, % L19 decreasong lightness, increasing chroma maps. % Cleared out some old experimental maps. % November 2010 Added R4, C6 and C7, reordered the naming of some cyclic maps. function [map, name, desc] = cmap(varargin) [I, N, chromaK, shift, reverse, diagnostics] = parseinputs(varargin{:}); % Default parameters for colour map construction. % Individual colour map specifications may override some of these. colourspace = 'LAB'; % Control points specified in CIELAB space sigma = 0; % Default smoothing for colour map lightness equalisation. splineorder = 3; % Default order of b-spline colour map curve. formula = 'CIE76'; % Default colour difference formula. W = [1 0 0]; % Colour difference weighting for Lightness, % chroma and hue (default is to only use Lightness) desc = ''; name = ''; attributeStr = ''; hueStr = ''; % Definitions of key angles in Lab space for defining colours in colour % blind colour spaces. Angles are in degrees. yellow575 = 92.62; % For protanopic and deuteranopic colour space. blue475 = -79.27; red660 = 32.36; % For tritanopic colour space. cyan485 = -138.73; switch I % Note all case expressions must be upper case. %----------------------------------------------------------------------------- %% Linear series case {'L1', 'L01', 'GREY', 'GRAY'} % Grey 0 - 100 desc = 'Grey scale'; attributeStr = 'linear'; hueStr = 'grey'; colpts = [ 0 0 0 100 0 0]; splineorder = 2; case {'L2', 'L02', 'REDUCEDGREY'} % Grey 10 - 95 desc = ['Grey scale with slightly reduced contrast to '... 'avoid display saturation problems']; attributeStr = 'linear'; hueStr = 'grey'; colpts = [10 0 0 95 0 0]; splineorder = 2; % Heat map adapted from the classic definition using straight line segments from % black to red to yellow to white. Here the spline order set to 3 and % additional control points to get a similar colour map path but with the % corners at red and yellow rounded off. The best heat map out there! case {'L3', 'L03', 'KRYW', 'HEAT'} desc = 'Black-Red-Yellow-White heat colour map'; attributeStr = 'linear'; hueStr = 'kryw'; colourspace = 'RGB'; splineorder = 3; colpts = [0 0 0 .85 0 0 1 .15 0 1 .85 0 1 1 .15 1 1 1 ]; case {'L4', 'L04', 'KRY', 'HEATYELLOW'} desc = 'Black-Red-Yellow heat colour map'; attributeStr = 'linear'; hueStr = 'kry'; colourspace = 'RGB'; splineorder = 3; colpts = [0 0 0 .85 0 0 1 .15 0 1 1 0]; case {'L5', 'L05', 'KGY'} desc = 'Colour Map along the green edge of CIELAB space'; attributeStr = 'linear'; hueStr = 'kgy'; colpts = [ 5 -9 5 15 -23 20 25 -31 31 35 -39 39 45 -47 47 55 -55 55 65 -63 63 75 -71 71 85 -79 79 95 -38 90]; case {'L6', 'L06', 'KBC'} desc = 'Blue shades running vertically up the blue edge of CIELAB space'; attributeStr = 'linear'; hueStr = 'kbc'; colpts = [ 5 31 -45 15 50 -66 25 65 -90 35 70 -100 45 45 -85 55 20 -70 65 0 -53 75 -22 -37 85 -38 -20 95 -25 -3]; case {'L7', 'L07', 'BMW'} desc = 'Blue-Pink-Light Pink colour map'; attributeStr = 'linear'; hueStr = 'bmw'; colpts = [ 5 29 -43 15 48 -66 25 64 -89 35 73 -100 45 81 -86 55 90 -69 65 83 -53 75 56 -36 85 32 -22 95 10 -7]; case {'L8', 'L08', 'BMY'} desc = 'Blue-Magenta-Yellow highly saturated colour map'; attributeStr = 'linear'; hueStr = 'bmy'; colpts = [10 ch2ab( 55,-58) 20 ch2ab( 75,-58) 30 ch2ab( 75,-40) 40 ch2ab( 73,-20) 50 ch2ab( 75, 0) 60 ch2ab( 70, 30) 70 ch2ab( 65, 60) 80 ch2ab( 75, 80) 95 ch2ab( 80, 105)]; case {'L9', 'L09', 'BGYW'} % Blue - green - yellow - white desc = 'Blue-green-yellow colour map'; attributeStr = 'linear'; hueStr = 'bgyw'; colpts = [20 59 -80 35 28 -66 45 -14 -29 60 -62 60 85 -10 85 95 -15 70 98 0 0]; case {'L10', 'GEOGRAPHIC'} desc = ['A "geographical" colour map. '... 'Best used with relief shading']; attributeStr = 'linear'; hueStr = 'gow'; colpts = [60 ch2ab(20, 180) % pale blue green 65 ch2ab(30, 135) 70 ch2ab(35, 75) 75 ch2ab(45, 85) 80 ch2ab(22, 90) 85 0 0 ]; case {'L11', 'GEOGRAPHIC2'} % Lighter version of L10 with a bit more chroma desc = ['A lighter "geographical" colour map. '... 'Best used with relief shading']; attributeStr = 'linear'; hueStr = 'gow'; colpts = [65 ch2ab(50, 135) % pale blue green 75 ch2ab(45, 75) 80 ch2ab(45, 85) 85 ch2ab(22, 90) 90 0 0 ]; case {'L11A', 'GEOGRAPHIC3'} % Lighter version of L11 desc = ['A lighter "geographical" colour map. '... 'Best used with relief shading']; attributeStr = 'linear'; hueStr = 'gow'; colpts = [70 ch2ab(50, 135) % pale blue green 80 ch2ab(45, 75) 85 ch2ab(45, 85) 90 ch2ab(22, 90) 95 0 0 ]; case {'L12', 'DEPTH'} desc = 'A "water depth" colour map'; attributeStr = 'linear'; hueStr = 'blue'; colpts = [95 0 0 80 ch2ab(20, -95) 70 ch2ab(25, -95) 60 ch2ab(25, -95) 50 ch2ab(35, -95)]; % The following three colour maps are for ternary images, eg Landsat images % and radiometric images. These colours form the nominal red, green and % blue 'basis colours' that are used to form the composite image. They are % designed so that they, and their secondary colours, have nearly the same % lightness levels and comparable chroma. This provides consistent feature % salience no matter what channel-colour assignment is made. The colour % maps are specified as straight lines in RGB space. For their derivation % see % https://arxiv.org/abs/1509.03700 case {'L13', 'REDTERNARY'} desc = 'red colour map for ternary images'; attributeStr = 'linear'; hueStr = 'ternary-red'; colourspace = 'RGB'; colpts = [0.00 0.00 0.00 0.90 0.17 0.00]; splineorder = 2; case {'L14', 'GREENTERNARY'} desc = 'green colour map for ternary images'; attributeStr = 'linear'; hueStr = 'ternary-green'; colourspace = 'RGB'; colpts = [0.00 0.00 0.00 0.00 0.50 0.00]; splineorder = 2; case {'L15', 'BLUETERNARY'} desc = 'blue colour map for ternary images'; attributeStr = 'linear'; hueStr = 'ternary-blue'; colourspace = 'RGB'; colpts = [0.00 0.00 0.00 0.10 0.33 1.00]; splineorder = 2; % Variation of L9 that starts at black. Works well. case {'L16', 'KBGYW'} % Black - Blue - green - yellow - white desc = 'Black-Blue-Green-Yellow-White colour map'; attributeStr = 'linear'; hueStr = 'kbgyw'; colpts = [10 0 0 20 59 -80 35 28 -66 45 -14 -29 60 -62 60 85 -10 85 95 -15 70 98 0 0]; W = [1,0,0]; % Linear decreasing lightness, increasing chroma to blue. case {'L17', 'WORB'} desc = 'White-Orange-Red-Blue, decreasing lightness with increasing saturation'; attributeStr = 'linear'; hueStr = 'worb'; % Construct a spiral of increasing chroma down through Lab space nsteps = 12; ang1 = 120; ang2 = -60; % Linearly interpolate hue angle ang = ang1:(ang2 - ang1)/(nsteps-1):ang2; % Interpolate chroma but use a 'gamma' of 0.5 to keep the colours more saturated. sat1 = 0; sat2 = 80; sat = ([sat1:(sat2-sat1)/(nsteps-1):sat2]/sat2).^0.5 * sat2; l1 = 100; l2 = 25; % Linearly interpolate lightness l = l1:(l2-l1)/(nsteps-1):l2; colpts = zeros(nsteps,3); for n=1:nsteps colpts(n,:) = [l(n), ch2ab(sat(n), ang(n))]; end splineorder = 2; case {'L18', 'WYOR'} desc = 'White-Yellow-Orange-Red, decreasing lightness with increasing saturation'; attributeStr = 'linear'; hueStr = 'wyor'; % Construct a spiral of increasing chroma down through Lab space nsteps = 12; ang1 = 120; ang2 = 35; % Linearly interpolate hue angle ang = ang1:(ang2 - ang1)/(nsteps-1):ang2; % Interpolate chroma but use a 'gamma' of 0.5 to keep the colours more saturated. sat1 = 0; sat2 = 84; sat = ([sat1:(sat2-sat1)/(nsteps-1):sat2]/sat2).^0.5 * sat2; l1 = 100; l2 = 45; % Linearly interpolate lightness l = l1:(l2-l1)/(nsteps-1):l2; colpts = zeros(nsteps,3); for n=1:nsteps colpts(n,:) = [l(n), ch2ab(sat(n), ang(n))]; end splineorder = 2; case {'L19', 'WCMR'} desc = 'White-Cyan-Magenta-Red, decreasing lightness with increasing saturation'; attributeStr = 'linear'; hueStr = 'wcmr'; % Construct a spiral of increasing chroma down through Lab % space. Use linear interolation for hue angle, chroma and lightness nsteps = 12; ang1 = -140; ang2 = 40; ang = ang1:(ang2 - ang1)/(nsteps-1):ang2; sat1 = 0; sat2 = 84; sat = sat1:(sat2-sat1)/(nsteps-1):sat2; l1 = 100; l2 = 45; l = l1:(l2-l1)/(nsteps-1):l2; colpts = zeros(nsteps,3); for n=1:nsteps colpts(n,:) = [l(n), ch2ab(sat(n), ang(n))]; end splineorder = 2; % A map inspired by Parula but with a smoother path that does not attempt to % include cyan and maintains a more uniform slope upwards in LAB % space. It also starts at a lightness of 20 with zero chroma. It works % very much better than Parula! case {'L20', 'GOULDIAN'} desc = 'Black-Blue-Green-Orange-Yellow map'; attributeStr = 'linear'; hueStr = 'kbgoy'; colpts = [20 0 0 40 55 -90 55 -47 0 70 -20 70 80 20 80 95 -21 92]; sigma = 0; splineorder = 3; %------------------------------------------------------------------------- %% Diverging colour maps % Note that on these colour maps often we do not go to full white but use a % lightness value of 95. This helps avoid saturation problems on monitors. % A lightness smoothing sigma of 5 to 7 is used to avoid generating a false % feature at the white point in the middle. Note however, this does create % a small perceptual contrast blind spot at the middle. case {'D1', 'D01', 'BWR', 'COOLWARM'} desc = ['Classic diverging blue - white - red colour map. End colours are' ... ' matched in lightness and chroma']; attributeStr = 'diverging'; hueStr = 'bwr'; colpts = [40 ch2ab(83,-64) 95 0 0 40 ch2ab(83, 39)]; sigma = 7; splineorder = 2; case {'D1A', 'D01A', 'BWRA'} desc = ['Diverging blue - white - red colour map. Similar to D1 but with darker'... ' end point colours']; attributeStr = 'diverging'; hueStr = 'bwr'; colpts = [20 ch2ab(49,-64) 40 ch2ab(83,-64) 65 ch2ab(60,-64) 95 0 0 65 ch2ab(60, 37) 40 ch2ab(83, 37) 20 ch2ab(49, 37)]; sigma = 7; splineorder = 2; case {'D1B', 'D01B', 'BWO'} desc = 'blue - white - dark orange colour map'; attributeStr = 'diverging'; hueStr = 'bwr'; colpts = [40 ch2ab(50,-64) 95 0 0 40 ch2ab(50, 77)]; sigma = 7; splineorder = 2; case {'D2', 'D02', 'GWV'} desc = 'Diverging green - white - violet colour map'; attributeStr = 'diverging'; hueStr = 'gwv'; colpts = [55 -50 55 95 0 0 55 60 -55]; sigma = 7; splineorder = 2; case {'D3', 'D03', 'GWR'} desc = 'Diverging green - white - red colour map'; attributeStr = 'diverging'; hueStr = 'gwr'; colpts = [55 -50 55 95 0 0 55 63 39]; sigma = 7; splineorder = 2; case {'D4', 'D04', 'BKR'} desc = 'Diverging blue - black - red colour map'; attributeStr = 'diverging'; hueStr = 'bkr'; colpts = [55 ch2ab(70, -76) 10 0 0 55 ch2ab(70, 35)]; sigma = 7; splineorder = 2; case {'D5', 'D05', 'GKR'} desc = 'Diverging green - black - red colour map'; attributeStr = 'diverging'; hueStr = 'gkr'; colpts = [60 ch2ab(80, 134) 10 0 0 60 ch2ab(80, 40)]; sigma = 7; splineorder = 2; case {'D6', 'D06', 'BKY'} desc = 'Diverging blue - black - yellow colour map'; attributeStr = 'diverging'; hueStr = 'bky'; colpts = [60 ch2ab(60, -85) 10 0 0 60 ch2ab(60, 85)]; sigma = 7; splineorder = 2; case {'D7', 'D07', 'BJY', 'DIVBJY'} % Linear diverging blue - grey - yellow. Works well desc = ['Linear-diverging blue - grey - yellow colour map. This kind' ... ' of diverging map has no perceptual dead spot at the centre']; attributeStr = 'diverging-linear'; hueStr = 'bjy'; colpts = [30 ch2ab(89, -59) 60 0 0 90 ch2ab(89, 96)]; splineorder = 2; case {'D8', 'D08', 'BJR'} % Linear diverging blue - grey - red desc = 'Linear-diverging blue - grey - red'; attributeStr = 'diverging-linear'; hueStr = 'bjr'; colpts = [30 ch2ab(105, -58) 42.5 0 0 55 ch2ab(105,41)]; splineorder = 2; case {'D9', 'D09'} % Lightened version of D1 for relief shading - Good. desc = ['Diverging low contrast blue - white - red colour map. Good in' ... ' conjunction with relief shading']; attributeStr = 'diverging'; hueStr = 'bwr'; colpts = [55 ch2ab(73,-74) 98 0 0 55 ch2ab(73, 39)]; sigma = 2; % Less smoothing needed for low contrast splineorder = 2; case 'D10' % low contrast diverging map for when you want to use % relief shading desc = ['Diverging low contrast cyan - white - magenta colour map. Good in' ... ' conjunction with relief shading']; attributeStr = 'diverging'; hueStr = 'cwm'; colpts = [80 ch2ab(44, -135) 100 0 0 80 ch2ab(44, -30)]; sigma = 0; % No smoothing needed for lightness range of 20 splineorder = 2; W = [1 1 1]; case 'D11' % Constant lightness diverging map for when you want to use % relief shading ? Perhaps lighten the grey so it is not quite % isoluminant ? desc = 'Diverging isoluminant lightblue - lightgrey - orange colour map'; attributeStr = 'diverging-isoluminant'; hueStr = 'cjo'; colpts = [70 ch2ab(50, -105) 70 0 0 70 ch2ab(50, 45)]; sigma = 7; splineorder = 2; W = [1 1 1]; case 'D12' % Constant lightness diverging map for when you want to use % relief shading ? Perhaps lighten the grey so it is not quite % isoluminant ? desc = 'Diverging isoluminant lightblue - lightgrey - pink colour map'; attributeStr = 'diverging-isoluminant'; hueStr = 'cjm'; colpts = [75 ch2ab(46, -122) 75 0 0 75 ch2ab(46, -30)]; sigma = 7; splineorder = 2; W = [1 1 1]; case {'D13', 'BWG'} % Pleasing Blue-White-Green diverging map desc = 'Diverging blue - white - green colour map'; attributeStr = 'diverging'; hueStr = 'bwg'; colpts = [20 ch2ab(40, -70) 45 ch2ab(60, -70) 70 ch2ab(50, -70-35) 95 0 0 95 0 0 70 ch2ab(50, 138+35) 45 ch2ab(60, 138) 20 ch2ab(40, 138)]; sigma = 5; splineorder = 3; %------------------------------------------------------------------------- %% Cyclic colour maps case {'C1', 'C01'} % I think this is my best zigzag style cyclic map - Good! % Control points are placed so that lightness steps up and % down are equalised. Additional intermediate points are % placed to try to even up the 'spread' of the key colours. desc = ['Cyclic: magenta - red - yellow - blue. Alows four' ... ' orientations/phase angles to be visulaised.']; attributeStr = 'cyclic'; hueStr = 'mrybm'; mag = [75 60 -40]; yel = [75 0 77]; blu = [35 70 -100]; red = [35 60 48]; colpts = [mag 55 70 0 red 55 35 62 yel 50 -20 -30 blu 55 45 -67 mag]; sigma = 7; splineorder = 2; % linear path case {'C2', 'C02', 'PHASE4'} % A big diamond across the gamut. Really good! Incorporates two % extra cotnrol points around blue to extend the width of that % segment slightly. desc = ['Cyclic: magenta - yellow - green - blue. Alows four' ... ' orientations/phase angles to be visualised.']; attributeStr = 'cyclic'; hueStr = 'mygbm'; colpts = [62.5 83 -54 80 20 25 95 -20 90 62.5 -65 62 42 10 -50 30 75 -103 48 70 -80 62.5 83 -54]; sigma = 7; splineorder = 2; case {'C2A', 'C02A'} % Attempt at a brighter version of C2. Not entirely successful desc = ''; attributeStr = 'cyclic'; hueStr = 'mygbm'; colpts = [71 75 -47 % magenta 97 -22 94 % yellow 71 -71 68 % green 45 41 -88 % blue 45 41 -88 71 75 -47]; sigma = 7; splineorder = 2; case {'C3', 'C03'} % white-red-black-blue-white - variation of C4 desc = ['Cyclic: white - red - black - blue']; attributeStr = 'cyclic'; hueStr = 'wrkbw'; colpts = [90 0 0 50 65 56 10 0 0 50 31 -80 90 0 0]; sigma = 7; splineorder = 2; case {'C4', 'C04', 'PHASE2'} % white-red-white-blue-white Works nicely desc = ['Cyclic: white - red - white - blue. Good if you just want to' ... ' visualise +ve and -ve phase']; attributeStr = 'cyclic'; hueStr = 'wrwbw'; colpts = [90 0 0 40 65 56 90 0 0 40 31 -80 90 0 0]; sigma = 7; splineorder = 2; case {'C5', 'C05', 'CYCLICGREY'} % Cyclic greyscale Works well desc = 'Cyclic: greyscale'; attributeStr = 'cyclic'; hueStr = 'grey'; colpts = [50 0 0 85 0 0 15 0 0 50 0 0]; sigma = 7; splineorder = 2; % A six colour map. Yellow, Cyan and Magenta at a lightness of 90. (Magenta gets % washed out with little chroma at 90) Red, Green and Blue at a lightness % of 50. The Green is a bit dark and the blue a bit light at 50. Overall a % slightly strange colour circle but I think it works quite well. case {'C6', 'C06'} desc = 'Six colour cyclic with primaries and secondaries matched in lightness.'; attributeStr = 'cyclic'; hueStr = 'rygcbmr'; y90 = [90 -7 90]; m90 = [90 24 -17]; c90 = [90 -48 -14]; y85 = [85 -22 84]; m85 = [85 35 -23]; c85 = [85 -46 -14]; y75 = [75 -16 74]; m75 = [75 60 -40]; c75 = [75 -39 -16]; r55 = [55 80 66]; b55 = [55 19 -71]; g55 = [55 -60 56]; r50 = [50 78 62]; b50 = [50 30 -81]; g50 = [50 -54 52]; colpts = [r50 y90 g50 c90 b50 m90 r50]; sigma = 7; splineorder = 2; case {'C7', 'C07'} % Zig-Zag Yellow - Magenta - Cyan - Green - Yellow % Colours are well balanced over the quadrants desc = 'Cyclic Yellow - Magenta - Cyan - Green - Yellow '; attributeStr = 'cyclic'; hueStr = 'ymcgy'; yel = [85 3 86]; y90 = [90 -20 90]; % a was -6 c90 = [90 -48 -14]; cyn = [85 -40 -25]; blu = [50 30 -79]; red = [50 79 64]; g70 = [70 -71 70]; g65 = [65 -67 66]; g60 = [60 -64 61]; g55 = [55 -60 57]; m70 = [70 75 -47]; m65 = [65 90 -55]; m60 = [60 98 -61]; m55 = [55 94 -71]; colpts = [y90 m60 c90 g60 y90]; sigma = 5; splineorder = 2; case {'C8', 'C08'} % Elliptical path - ok desc = 'Cyclic: low contrast colour ellipse'; attributeStr = 'cyclic'; hueStr = 'mygbm'; ang = 112; colpts = [70 ch2ab(46, ang-90) 90 ch2ab(82, ang) 70 ch2ab(46, ang+90) 50 ch2ab(82, ang+180) 70 ch2ab(46, ang-90)]; W = [1 1 1]; case {'C9', 'C09'} % Elliptical path. Greater range of lightness values and % slightly more saturated colours. Seems to work however I do % not find the colour sequence that attractive. This is a % constraint of the gamut. desc = 'Cyclic: colour ellipse'; attributeStr = 'cyclic'; hueStr = 'mybm'; ang = 124; colpts = [60 ch2ab(40, ang-90) 100 ch2ab(98, ang) 60 ch2ab(40, ang+90) 20 ch2ab(98, ang+180) 60 ch2ab(40, ang-90)]; W = [1 1 1]; sigma = 7; case 'C10' % Circle at 67 - sort of ok but a bit flouro desc = 'Cyclic: isoluminant'; attributeStr = 'cyclic-isoluminant colour circle'; hueStr = 'mgbm'; chr = 42; ang = 124; colpts = [67 ch2ab(chr, ang-90) 67 ch2ab(chr, ang) 67 ch2ab(chr, ang+90) 67 ch2ab(chr, ang+180) 67 ch2ab(chr, ang-90)]; W = [1 1 1]; case 'C11' % Variation of C1. Perceptually this is good. Excellent balance % of colours in the quadrants but the colour mix is not to my % taste. Don't like the green. The red-green transition % clashes desc = ['Cyclic: blue - green - red - magenta. Allows four' ... ' orientations/phase angles to be visulaised.']; attributeStr = 'cyclic'; hueStr = 'bgrmb'; blu = [35 70 -100]; colpts = [blu 70 -70 64 35 65 50 70 75 -46 blu ]; sigma = 7; splineorder = 2; % linear path %----------------------------------------------------------------------------- %% Rainbow style colour maps case {'R1', 'R01', 'RAINBOW'} % Reasonable rainbow colour map after it has been % fixed by equalisecolourmap. desc = ['The least worst rainbow colour map I can devise. Note there are' ... ' small perceptual blind spots at yellow and red']; attributeStr = 'rainbow'; hueStr = 'bgyrm'; colpts = [35 63 -98 45 -14 -30 60 -55 60 85 0 80 55 60 62 75 55 -35]; sigma = 7; splineorder = 2; % linear path case {'R2', 'R02', 'RAINBOW2'} % Similar to R1 but with the colour map finishing % at red rather than continuing onto pink. desc = ['Reasonable rainbow colour map from blue to red. Note there is' ... ' a small perceptual blind spot at yellow']; attributeStr = 'rainbow'; hueStr = 'bgyr'; colpts = [35 60 -98 45 -15 -30 60 -55 60 85 0 78 55 73 68]; sigma = 5; splineorder = 2; % linear path case {'R2A', 'R02A'} % 'Light', low contrast rainbow map for use with relief shading desc = 'Light, low contrast rainbow map for use with relief shading'; attributeStr = 'rainbow'; hueStr = 'bgyrm'; colpts = [50 30 -80 60 -40 0 85 0 78 60 60 60 70 60 -20 80 40 -30]; sigma = 7; splineorder = 2; % linear path W = [1,0,0]; case {'R2AA', 'R02AA'} % 'Light', low contrast rainbow map for use with relief shading desc = 'Light, low contrast rainbow map for use with relief shading'; attributeStr = 'rainbow'; hueStr = 'bgyrm'; colpts = [50 30 -80 60 -40 0 85 0 78 60 60 60 70 60 -20]; sigma = 7; splineorder = 2; % linear path W = [1,0,0]; case {'R3', 'R03', 'RAINBOW3'} % Diverging rainbow. The blue and red % points are matched in lightness and % chroma as are the green and magenta % points. Works well. desc = ['Diverging-rainbow colourmap. Yellow is the central reference' ... ' colour. The blue and red end points are matched in lightness' ... ' and chroma']; attributeStr = 'diverging-rainbow'; hueStr = 'bgymr'; colpts = [45 39 -83 52 -23 -23 60 -55 55 85 -2 85 60 74 -17 45 70 59]; sigma = 5; splineorder = 2; % linear path % More vivid version of R2 with greater range of luminance. However I % find the vivid colours tend to make you segment regions in your data % on the basis of colour and you can lose the overall sense of structure. case {'R4', 'R04', 'RAINBOW4'} desc = ['Rainbow colour map from blue to red. Note there is' ... ' a small perceptual blind spot at yellow']; attributeStr = 'rainbow'; hueStr = 'bgyr'; colpts = [10 42 -57 20 59 -80 40 55 -93 50 -23 -25 60 -60 60 90 -12 89 70 33 72 55 75 65 45 70 55]; sigma = 5; splineorder = 2; % linear path %----------------------------------------------------------------------------- %% Isoluminant colour maps case {'I1', 'I01'} desc = ['Isoluminant blue to green to orange at lightness 70. '... 'Poor on its own but works well with relief shading']; attributeStr = 'isoluminant'; hueStr = 'cgo'; colpts = [70 ch2ab(40, -115) 70 ch2ab(50, 160) 70 ch2ab(50, 90) 70 ch2ab(50, 45)]; W = [1 1 1]; case {'I2', 'I02'} % Adaptation of I1 shifted to 80 from 70 desc = ['Isoluminant blue to green to orange at lightness 80. '... 'Poor on its own but works well with relief shading']; attributeStr = 'isoluminant'; hueStr = 'cgo'; colpts = [80 ch2ab(36, -115) 80 ch2ab(50, 160) 80 ch2ab(50, 90) 80 ch2ab(46, 55)]; W = [1 1 1]; case {'I3', 'I03'} desc = ['Isoluminant blue to pink at lightness 70. '... 'Poor on its own but works well with relief shading']; attributeStr = 'isoluminant'; hueStr = 'cm'; colpts = [70 ch2ab(40, -125) 70 ch2ab(40, -80) 70 ch2ab(40, -40) 70 ch2ab(50, 0)]; W = [1 1 1]; %----------------------------------------------------------------------------- %% Colour Maps for the Colour Blind %% Protanopic / Deuteranopic colour maps % The Protanopic / Deuteranopic colour space is represented by two planes. % One plane is defined by the neutral axis and the colour point at yellow % 575nm, and the other defined by the neutral axis and the colour point at % blue 475nm. % Hue angles in a,b space of the key colours common to Protanopic/Deuteranopic % and Trichromatic viewers are: % yellow 575nm = 1.6165 radians, 92.62 degrees % blue 475nm = -1.3835 radians, -79.27 degrees case 'CBL1' % Linear/diverging map for Protanopic/Deuteranopic viewers The % symmetry requirements for diverging maps means that the % colours are not as saturated as one would like. However it % works better than CBL2 desc = 'Linear map for Protanopic/Deuteranopic viewers'; attributeStr = 'linear-protanopic-deuteranopic'; hueStr = 'kbjyw'; colpts = [5 0 0 28 ch2ab(61, blue475) 50 0 0 72 ch2ab(61, yellow575) 95 0 0]; W = [1 0 0]; splineorder = 3; case 'CBL2' % Linear map with maximal chroma for Protanopic/Deuteranopic % viewers. Does not work as well as one would hope. The % colours are too uneven. desc = 'Linear map with maximal chroma for Protanopic/Deuteranopic viewers'; attributeStr = 'linear-protanopic-deuteranopic'; hueStr = 'kbw'; colpts = [5 0 0 20 ch2ab(30, blue475) 58 ch2ab(69, blue475) 68 0 0 85 ch2ab(98, yellow575) 96 ch2ab(5, yellow575) 98 0 0]; W = [1 0 0]; splineorder = 2; sigma = 9; case 'CBL3' % Linear map up the blue edge of the colour space desc = 'Linear blue map for Protanopic/Deuteranopic viewers'; attributeStr = 'linear-protanopic-deuteranopic'; hueStr = 'kbw'; colpts = [5 0 0 56 ch2ab(68, blue475) 95 0 0]; W = [1 0 0]; splineorder = 2; sigma = 9; case 'CBL4' % Linear map up the yellow edge of the colour space desc = 'Linear yellow map for Protanopic/Deuteranopic viewers'; attributeStr = 'linear-protanopic-deuteranopic'; hueStr = 'kyw'; colpts = [5 0 0 25 ch2ab(34, yellow575) 88 ch2ab(88, yellow575) 95 ch2ab(5, yellow575)]; W = [1 0 0]; splineorder = 2; sigma = 9; case 'CBD1' % Diverging map blue-white-yellow desc = 'Diverging map blue-white-yellow'; attributeStr = 'diverging-protanopic-deuteranopic'; hueStr = 'bwy'; colpts = [60 ch2ab(63, blue475) 95 0 0 60 ch2ab(63, yellow575)]; W = [1 0 0]; splineorder = 2; sigma = 5; case 'CBD2' % Diverging-linear map blue-grey-yellow desc = 'Diverging-linear map blue-grey-yellow'; attributeStr = 'diverging-linear-protanopic-deuteranopic'; hueStr = 'bjy'; colpts = [57 ch2ab(67, blue475) 73 0 0 89 ch2ab(67, yellow575)]; W = [1 0 0]; splineorder = 2; case 'CBC1' % 4-phase cyclic map blue-white-yellow-black desc = '4-phase cyclic map blue-white-yellow-black'; attributeStr = 'cyclic-protanopic-deuteranopic'; hueStr = 'bwyk'; colpts = [56 ch2ab(61, blue475) 96 0 0 56 ch2ab(61, yellow575) 16 0 0]; W = [1 0 0]; splineorder = 2; sigma = 5; case 'CBC2' % 2-phase cyclic map white-yellow-white-blue desc = '2-phase cyclic map white-yellow-white-blue'; attributeStr = 'cyclic-protanopic-deuteranopic'; hueStr = 'wywb'; colpts = [96 0 0 55 ch2ab(68, yellow575) 96 0 0 55 ch2ab(68, blue475)]; W = [1 0 0]; splineorder = 2; sigma = 5; %% Tritanopic colour maps % Hue angles in a,b space of the key colours common to Tritanopic % and Trichromatic viewers. % red 660nm 0.5648 radians, 32.36 degrees % cyan 485nm -2.4213 radians, -138.73 degrees case 'CBTL1' % Tritanopic linear map with maximal chroma desc = 'Tritanopic linear map with maximal chroma'; attributeStr = 'linear-tritanopic'; hueStr = 'krjcw'; colpts = [5 0 0 20 ch2ab(50, red660) 56 ch2ab(95, red660) 78 ch2ab(52, cyan485) 98 0 0]; W = [1 0 0]; splineorder = 2; sigma = 9; case 'CBTL2' % Tritanopic linear map, can also be used as a diverging % map. However the symmetry requirements of a diverging % map results in colours of lower chroma desc = 'Tritanopic linear map'; attributeStr = 'linear-tritanopic'; hueStr = 'krjcw'; colpts = [5 0 0 25 ch2ab(58, red660) 50 0 0 75 ch2ab(58, cyan485) 95 0 0]; W = [1 0 0]; splineorder = 3; case 'CBTL3' % Linear map up the blue green edge of the colour space desc = 'Tritanopic linear blue map'; attributeStr = 'linear-tritanopic'; hueStr = 'kcw'; colpts = [5 0 0 70 ch2ab(40, cyan485) 85 ch2ab(40, cyan485) 95 0 0]; W = [1 0 0]; splineorder = 3; case 'CBTL4' % Linear map up the red edge of the colour space desc = 'Tritanopic linear red/heat map'; attributeStr = 'linear-tritanopic'; hueStr = 'krw'; colpts = [5 0 0 45 ch2ab(100, red660) 70 ch2ab(50, red660) 95 0 0]; W = [1 0 0]; splineorder = 3; case 'CBTD1' % Tritanopic diverging map desc = 'Tritanopic diverging map'; attributeStr = 'diverging-tritanopic'; hueStr = 'cwr'; colpts = [75 ch2ab(39, cyan485) 98 0 0 75 ch2ab(39, red660)]; W = [1 0 0]; splineorder = 2; sigma = 5; case 'CBTC1' % 4-phase tritanopic cyclic map desc = '4-phase tritanopic cyclic map'; attributeStr = 'cyclic-tritanopic'; hueStr = 'cwrk'; colpts = [70 ch2ab(39, cyan485) 100 0 0 70 ch2ab(39, red660) 40 0 0]; W = [1 0 0]; splineorder = 2; sigma = 5; case 'CBTC2' % 2-phase tritanopic cyclic map desc = '2-phase tritanopic cyclic map'; attributeStr = 'cyclic-tritanopic'; hueStr = 'wrwc'; colpts = [100 0 0 70 ch2ab(41, red660) 100 0 0 70 ch2ab(41, cyan485)]; W = [1 0 0]; splineorder = 2; sigma = 5; %----------------------------------------------------------------------------- %% Experimental colour maps and colour maps that illustrate some design principles case 'X1' desc = ['Two linear segments with different slope to illustrate importance' ... ' of lightness gradient.']; attributeStr = 'linear-lightnessnormalised'; hueStr = 'by'; colpts = [30 ch2ab(102, -54) 40 0 0 90 ch2ab(90, 95)]; W = [1 0 0]; splineorder = 2; case 'X2' desc = ['Two linear segments with different slope to illustrate importance' ... ' of lightness gradient.']; attributeStr = 'linear-CIE76normalised'; hueStr = 'by'; colpts = [30 ch2ab(102, -54) 40 0 0 90 ch2ab(90, 95)]; W = [1 1 1]; splineorder = 2; case 'X3' % Constant lightness 'v' path to test unimportance of having a smooth % path in hue/chroma. Slight 'feature' at the red corner (Seems more % important on poor monitors) attributeStr = 'isoluminant-HueChromaSlopeDiscontinuity'; hueStr = 'brg'; colpts = [50 17 -78 50 77 57 50 -48 50]; splineorder = 2; % linear path W = [1 1 1]; % Yukky looking attempt at getting a constantly increasing lightness path case {'X5', 'X05'} desc = ''; attributeStr = 'rainbow'; hueStr = ''; b20 = [20 59 -80]; g40 = [40 -45 45]; b40 = [40 25 -65]; g50 = [50 -54 52]; r55 = [55 80 66]; o65 = [65 51 73]; g80 = [80 -80 77]; y90 = [90 -7 90]; y97 = [97 -22 94]; colpts = [b20 b40 r55 g80 y97]; sigma = 5; splineorder = 2; % linear path % Rainbow constructed from an unwinding of the C6 cyclic % map. Interesting but I don't think it is the best thing to do case {'X6', 'X06'} desc = 'Rainbow constructed from an unwinding of the C6 cyclic map' attributeStr = 'linear'; hueStr = 'bcgyrm'; y90 = [90 -7 90]; m90 = [90 24 -17]; c90 = [90 -48 -14]; y85 = [85 -22 84]; m85 = [85 35 -23]; c85 = [85 -46 -14]; y75 = [75 -16 74]; m75 = [75 60 -40]; c75 = [75 -39 -16]; r55 = [55 80 66]; b55 = [55 19 -71]; g55 = [55 -60 56]; r50 = [50 78 62]; b50 = [50 30 -81]; g50 = [50 -54 52]; colpts = [b50 c90 g50 y90 r50 m90]; sigma = 7; splineorder = 2; % A map inspired by Parula but with a smoother path that does not attempt to % include cyan and maintains a more uniform slope upwards in LAB space. % It works very much better than Parula! case 'X7' desc = 'Blue-Green-Orange-Yellow map similar to Parula but better!'; attributeStr = 'linear'; hueStr = 'bgoy'; colpts = [25 50 -70 40 35 -75 55 -47 0 70 -20 70 80 20 80 95 -21 92]; sigma = 0; splineorder = 3; % A set of isoluminant colour maps only varying in saturation to test % the importance of saturation (not much) Colour Maps are linear with % a reversal to test importance of continuity. case 'X10' % Isoluminant 50 only varying in saturation attributeStr = 'isoluminant'; hueStr = 'r'; colpts = [50 0 0 50 77 64 50 0 0]; splineorder = 2; sigma = 0; W = [1 1 1]; case 'X11' % Isoluminant 50 only varying in saturation attributeStr = 'isoluminant'; hueStr = 'b'; colpts = [50 0 0 50 0 -56 50 0 0]; splineorder = 2; sigma = 0; W = [1 1 1]; case 'X12' % Isoluminant 90 only varying in saturation attributeStr = 'isoluminant'; hueStr = 'isoluminant_90_g'; colpts = [90 0 0 90 -76 80 90 0 0]; splineorder = 2; sigma = 0; W = [1 1 1]; % Difference in CIE76 and CIEDE2000 in chroma case 'X13' % Isoluminant 55 only varying in chroma. CIEDE76 attributeStr= 'isoluminant-CIE76'; hueStr = 'jr'; colpts = [55 0 0 55 80 67]; splineorder = 2; W = [1 1 1]; formula = 'CIE76'; case 'X14' % Same as X13 but using CIEDE2000 attributeStr= 'isoluminant-CIEDE2000'; hueStr = 'jr'; colpts = [55 0 0 55 80 67]; splineorder = 2; W = [1 1 1]; formula = 'CIEDE2000'; case 'X15' % Grey 0 - 100. Same as No 1 but with CIEDE2000 desc = 'Grey scale'; attributeStr= 'linear-CIEDE2000'; hueStr = 'grey'; colpts = [ 0 0 0 100 0 0]; splineorder = 2; formula = 'CIEDE2000'; case 'X16' % Isoluminant 30 only varying in chroma attributeStr= 'isoluminant'; hueStr = 'b'; colpts = [30 0 0 30 77 -106]; splineorder = 2; W = [1 1 1]; case 'X21' % Blue to yellow section of rainbow map R1 for illustrating % colour ordering issues attributeStr= 'rainbow-section1'; hueStr = 'bgy'; colpts = [35 60 -100 45 -15 -30 60 -55 60 85 0 80]; splineorder = 2; % linear path case 'X22' % Red to yellow section of rainbow map R1 for illustrating % colour ordering issues attributeStr= 'rainbow-section2'; hueStr = 'ry'; colpts = [55 70 65 85 0 80]; splineorder = 2; % linear path case 'X23' % Red to pink section of rainbow map R1 for illustrating % colour ordering issues attributeStr= 'rainbow-section3'; hueStr = 'rm'; colpts = [55 70 65 75 55 -35]; splineorder = 2; % linear path case 'XD1' % Same as D1 but with no smoothing desc = 'Diverging blue-white-red colour map'; attributeStr = 'diverging'; hueStr = 'bwr'; colpts = [40 ch2ab(83,-64) 95 0 0 40 ch2ab(83, 39)]; sigma = 0; splineorder = 2; case 'X30' desc = 'red - green - blue interpolated in rgb'; attributeStr = 'linear'; hueStr = 'rgb'; colourspace = 'RGB'; colpts = [1.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 1.00]; sigma = 0; W = [0 0 0]; splineorder = 2; case 'X31' desc = 'red - green - blue interpolated in CIELAB'; attributeStr = 'linear'; hueStr = 'rgb'; colourspace = 'LAB'; colpts = [53 80 67 88 -86 83 32 79 -108]; sigma = 0; W = [0 0 0]; splineorder = 2; case 'XI4' % Maxiumum chroma isoluminant map at lightness 65 colpts = [65 ch2ab(90, 135) 65 ch2ab(88, 53) 65 ch2ab(60, 10) 65 ch2ab(90, -32)]; W = [1 1 1]; splineorder = 2; sigma = 9; case 'XI5' % Maxiumum chroma isoluminant map at lightness 65 colpts = [65 ch2ab(90, 135) 65 ch2ab(90, -32)]; W = [1 1 1]; splineorder = 2; case 'XI6' % Maxiumum chroma isoluminant map for deuteranope at % lightness 60 yellow575 = 92.62; % degrees blue475 = -79.27; % degrees colpts = [60 ch2ab(64, blue475) 60 ch2ab(64, yellow575)]; W = [1 1 1]; splineorder = 2; case 'XD7A' % Linear diverging blue - magenta- grey - orange - yellow. % Modified from 'D7' to have a double arch shaped path in an attempt % to improve its Metric properties. Also starts at lightness % of 40 rather than 30. The centre grey region is a bit too % prominant and overall the map is perhaps a bit too 'bright' attributeStr = 'diverging-linear'; hueStr = 'bmjoy'; colpts = [40 ch2ab(88, -64) 55 ch2ab(70, -30) 64 ch2ab(2.5, -72.5) 65 0 0 66 ch2ab(2.5, 107.5) 75 ch2ab(70, 70) 90 ch2ab(88,100)]; splineorder = 3; case 'XD7C' % Linear diverging green - grey - yellow % Reasonable, perhaps easier on the eye than D7 attributeStr = 'diverging-linear'; hueStr = 'gjy'; rad = 65; colpts = [40 ch2ab(rad, 136) 65 0 0 90 ch2ab(rad, 95)]; splineorder = 2; case 'XL102' % Linear decreasing lightness, increasing saturation to blue % Try to add intermediate colours YUK! attributeStr = 'linear'; huestr = 'wb'; colpts = [90 0 0 58 ch2ab(57, 130) 40 ch2ab(85, 45) 25 ch2ab(114, -54)]; splineorder = 2; case 'XL103' % Linear increasing lightness, increasing saturation to yellow % Perhpaps the least objectionable increasing lightness % and saturation colour combination attributeStr = 'linear'; huestr = 'km'; colpts = [10 0 0 97 -22 94]; splineorder = 2; case 'XL104' % Linear increasing lightness, increasing saturation to magenta attributeStr = 'linear'; huestr = 'km'; colpts = [10 0 0 70 74 -46]; splineorder = 2; %%------------------------------------------------------------- otherwise % Invoke the catalogue search to help the user catalogue(I); clear map; clear name; clear desc; return end % Adjust chroma/saturation but only if colourspace is LAB if strcmpi(colourspace, 'LAB') colpts(:,2:3) = chromaK * colpts(:,2:3); end % Colour map path is formed via a b-spline in the specified colour space Npts = size(colpts,1); if Npts < 2 error('Number of input points must be 2 or more') elseif Npts < splineorder splineorder = Npts; fprintf('Warning: Spline order is greater than number of data points\n') fprintf('Reducing order of spline to %d\n', Npts) end % Rely on the attribute string to identify if colour map is cyclic. We may % want to construct a colour map that has identical endpoints but do not % necessarily want continuity in the slope of the colour map path. if strfind(attributeStr, 'cyclic') cyclic = 1; labspline = pbspline(colpts', splineorder, N); else cyclic = 0; labspline = bbspline(colpts', splineorder, N); end % Apply contrast equalisation with required parameters. Note that sigma is % normalised with respect to a colour map of length 256 so that if a short % colour map is specified the smoothing that is applied is adjusted to suit. sigma = sigma*N/256; map = equalisecolourmap(colourspace, labspline', formula,... W, sigma, cyclic, diagnostics); % If specified apply a cyclic shift to the colour map if shift if isempty(strfind(attributeStr, 'cyclic')) fprintf('Warning: Colour map shifting being applied to a non-cyclic map\n'); end map = circshift(map, round(N*shift)); end if reverse map = flipud(map); end % Compute mean chroma of colour map lab = rgb2lab(map); meanchroma = sum(sqrt(sum(lab(:,2:3).^2, 2)))/N; % Construct lightness range description if strcmpi(colourspace, 'LAB') % Use the control points L = colpts(:,1); else % For RGB use the converted CIELAB values L = round(lab(:,1)); end minL = min(L); maxL = max(L); if minL == maxL % Isoluminant LStr = sprintf('%d', minL); elseif L(1) == maxL && ... (~isempty(strfind(attributeStr, 'diverging')) ||... ~isempty(strfind(attributeStr, 'linear'))) LStr = sprintf('%d-%d', maxL, minL); else LStr = sprintf('%d-%d', minL, maxL); end % Build overall colour map name name = sprintf('%s_%s_%s_c%d_n%d',... attributeStr, hueStr, LStr, round(meanchroma), N); if shift name = sprintf('%s_s%d', name, round(shift*100)); end if reverse name = sprintf('%s_r', name); end if diagnostics % Print description and plot path in colourspace fprintf('%s\n',desc); colourmappath(map, 'fig', 10) end %------------------------------------------------------------------ % Conversion from (chroma, hue angle) description to (a*, b*) coords function ab = ch2ab(chroma, angle_degrees) theta = angle_degrees/180*pi; ab = chroma*[cos(theta) sin(theta)]; %------------------------------------------------------------------ % % Function to list colour maps with names containing a specific string. % Typically this is used to search for colour maps having a specified attribute: % 'linear', 'diverging', 'rainbow', 'cyclic', 'isoluminant' or 'all'. % This code is awful! function catalogue(str) if ~exist('str', 'var') str = 'all'; end % Get all case expressions in this function caseexpr = findcaseexpr; % Construct all colour map names for n = 1:length(caseexpr) % Get 1st comma separated element in caseexpr label = strtok(caseexpr{n},','); [~, name{n}] = cmap(label); end % Check each colour name for the specified search string. Exclude the % experimental maps with label starting with X. fprintf('\n CMAP label(s) Colour Map name\n') fprintf('------------------------------------------------\n') found = 0; for n = 1:length(caseexpr) if caseexpr{n}(1) ~= 'X' if any(strfind(upper(name{n}), str)) || strcmpi(str, 'all') fprintf('%-20s %s\n', caseexpr{n}, name{n}); found = 1; end end end if ~found fprintf('Sorry, no colour map with label or attribute %s found\n', str); end %----------------------------------------------------------------------- % Function to find all the case expressions in this function. Yuk there must % be a better way! Assumes the case statement do not span more than one line function caseexpr = findcaseexpr [fid, msg] = fopen([mfilename('fullpath') '.m'], 'r'); error(msg); caseexpr = {}; n = 0; line = fgetl(fid); while ischar(line) [tok, remain] = strtok(line); if strcmpi(tok, 'case') % If we are here remain should be the case expression. % Remove any trailing comment from case expression [tok, remain] = strtok(remain,'%'); % Remove any curly brackets from the case expression tok(tok=='{') = []; tok(tok=='}') = []; tok(tok=='''') = []; n = n+1; caseexpr{n} = tok; end line = fgetl(fid); end caseexpr = strtrim(caseexpr); fclose(fid); %----------------------------------------------------------------------- % Function to parse the input arguments and set defaults function [I, N, chromaK, shift, reverse, diagnostics] = parseinputs(varargin) p = inputParser; numericORchar = @(x) isnumeric(x) || ischar(x); numericORlogical = @(x) isnumeric(x) || islogical(x); % The first argument is either a colour map label string or a string to % search for in a colourmap name. If no argument is supplied it is assumed % the user wants to list all possible colourmaps. addOptional(p, 'I', 'all', @ischar); % Optional parameter-value pairs and their defaults % ** Note if you are using R2012 or earlier you should change the calls to % 'addParameter' below to 'addParamValue' and hopefully the code will work % for you. addParameter(p, 'N', 256, @isnumeric); addParameter(p, 'shift', 0, @isnumeric); addParameter(p, 'chromaK', 1, @isnumeric); addParameter(p, 'reverse', 0, numericORlogical); addParameter(p, 'diagnostics', 0, numericORlogical); parse(p, varargin{:}); I = strtrim(upper(p.Results.I)); N = p.Results.N; chromaK = p.Results.chromaK; shift = p.Results.shift; reverse = p.Results.reverse; diagnostics = p.Results.diagnostics; if abs(shift) > 1 error('Cyclic shift fraction magnitude cannot be larger than 1'); end if chromaK < 0 error('chromaK must be greater than 0') end if chromaK > 1 fprintf('Warning: chromaK is greater than 1. Gamut clipping may occur') end