WebSVN – seema-scanner – Diff – /matlab/alignSubScansMarkers.m


function MSE = alignSubScansMarkers(calibrationFileName, alnFileName)
%ALIGNSUBSCANSMARKERS Determines an exact alignment of sub scans (scans
% from e.g. one revolution of the rotation stage). 
% The method searches for circular white markers of a specific diameter.
% White frames corresponding to each sub scan must be available.
% A coarse alignment in the form of an aln-file must be provided. 
%
% 2017 Jakob Wilm, DTU
 
initialAlign = readMeshLabALN(alnFileName);
[alnFilePath, ~, ~] = fileparts(alnFileName);
 
calibration = readOpenCVXML(calibrationFileName);
 
% correct for Matlab 1-indexing in principle point coordinates
calibration.K0(1:2, 3) = calibration.K0(1:2, 3)+1;
calibration.K1(1:2, 3) = calibration.K1(1:2, 3)+1;
 
% full projection matrices in Matlab convention
P0 = transpose(calibration.K0*[eye(3) zeros(3,1)]);
P1 = transpose(calibration.K1*[calibration.R1 calibration.T1']);
 
% matlab cam params for undistortion
camParams0 = cameraParameters('IntrinsicMatrix', calibration.K0', 'RadialDistortion', calibration.k0([1 2 5]), 'TangentialDistortion', calibration.k0([3 4]));
camParams1 = cameraParameters('IntrinsicMatrix', calibration.K1', 'RadialDistortion', calibration.k1([1 2 5]), 'TangentialDistortion', calibration.k1([3 4]));
 
% matlab struct for triangulation
camStereoParams = stereoParameters(camParams0, camParams1, calibration.R1', calibration.T1');
 
nSubScans = length(initialAlign);
 
% 3D coordinates of markers in local camera frame
E = cell(nSubScans, 1);
 
% 3D coordinates of markers in global initial alignment
Eg = cell(size(E));
 
% find 3D markers coordinates 
for i=1:nSubScans
%for i=5:5
    % load point cloud
    pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);
    pcFilePath = fileparts(pcFileName);
    pc = pcread(pcFileName);
    Q = pc.Location;
    idString = strsplit(initialAlign(i).FileName, {'.ply', '_'});
    idString = idString{end-1};
    
    % load white frames
    frame0 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames0_0.png'));
    frame1 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames1_0.png'));
 
    e0Coords = autoDetectMarkers(frame0);
    e1Coords = autoDetectMarkers(frame1);
    
    %e0Coords = manuallyDetectMarkers(frame0);
    %e1Coords = manuallyDetectMarkers(frame1);
    
    %[e0Coords, conf0] = detectMarkersSubpix(frame0, e0Coords, P0, Q);
    %[e1Coords, conf1] = detectMarkersSubpix(frame1, e1Coords, P1, Q);
 
    if(length(e0Coords) < 1 || length(e1Coords) < 1)
        continue;
    end
    
%     figure; 
%     subplot(1,2,1);
%     imshow(frame0);
%     hold('on');
%     plot(e0Coords(:,1), e0Coords(:,2), 'rx', 'MarkerSize', 15);
%     subplot(1,2,2);
%     imshow(frame1);
%     hold('on');
%     plot(e1Coords(:,1), e1Coords(:,2), 'rx', 'MarkerSize', 15);
%     drawnow;
    
    e0Coords = undistortPoints(e0Coords, camParams0);
    e1Coords = undistortPoints(e1Coords, camParams1);
 
    % match ellipse candidates between cameras based on projection
    E0 = projectOntoPointCloud(e0Coords, P0, Q);
    E1 = projectOntoPointCloud(e1Coords, P1, Q);
 
    matchedPairs = nan(size(E0, 1), 2);
    nMatchedPairs = 0;
    for j=1:size(E0, 1)
        
        % should use pdist2 instead..
        sqDists = sum((E1 - repmat(E0(j,:), size(E1, 1), 1)).^2, 2);
        
        [minSqDist, minSqDistIdx] = min(sqDists);
 
        if(minSqDist < 1^2)
            nMatchedPairs = nMatchedPairs + 1;
            matchedPairs(nMatchedPairs, :) = [j, minSqDistIdx];
        end
    end
    matchedPairs = matchedPairs(1:nMatchedPairs, :);
    
    figure; 
    subplot(1,2,1);
    imshow(frame0);
    hold('on');
    plot(e0Coords(matchedPairs(:, 1),1), e0Coords(matchedPairs(:, 1),2), 'rx', 'MarkerSize', 15);
    subplot(1,2,2);
    imshow(frame1);
    hold('on');
    plot(e1Coords(matchedPairs(:, 2),1), e1Coords(matchedPairs(:, 2),2), 'rx', 'MarkerSize', 15);
    drawnow;
    
%     % triangulate marker centers (lens correction has been performed)
%     [E{i}, e] = triangulate(e0Coords(matchedPairs(:, 1),:), e1Coords(matchedPairs(:, 2),:), camStereoParams);
%     E{i} = E{i}(e<3.0, :);
%     display(e);
    
    [E{i}, e] = detectMarkersStereoSubpix(frame0, frame1, E0(matchedPairs(:, 1), :), camStereoParams, pc);
    display(e);
    
    % write point cloud with marker (debugging)
    pcDebug = pointCloud([pc.Location; E{i}], 'Color', [pc.Color; repmat([255, 0, 0], size(E{i}, 1), 1)]);
    pcwrite(pcDebug, 'pcDebug.ply');
    
    % bring markers into initial alignment
    [U,~,V] = svd(initialAlign(i).Rotation);
    Ri = U*V';
    Ti = initialAlign(i).Translation;
    
    Eg{i} = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);
end
 
% show found markers in initial alignment
figure;
hold('on');
for i=1:nSubScans
    % fix Ri to be orthogonal
    [U,~,V] = svd(initialAlign(i).Rotation);
    Ri = U*V';
    
    % bring point cloud into initial alignment
    pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);
    pc = pcread(pcFileName);
    tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);
    pcg = pctransform(pc, tform);
   
    pcshow(pcg);
    xlabel('x');
    ylabel('y');
    zlabel('z');
    
    plot3(Eg{i}(:,1), Eg{i}(:,2), Eg{i}(:,3), '.', 'MarkerSize', 15);
    title('Initial Alignment');
end
 
% match markers between poses using initial alignment
Pg = {};
P = {};
for i=1:nSubScans
    for j=1:size(Eg{i}, 1)
        pg = Eg{i}(j,:);
        p = E{i}(j,:);
        matched = false;
        for k=1:size(Pg, 2)
            clusterCenter = mean(cat(1, Pg{:,k}), 1);
            if(sum((pg - clusterCenter).^2) < 3^2)
                % store in global frame
                Pg{i,k} = pg;
                % store in local frame
                P{i,k} = p;
                matched = true;
                break;
            end
        end
        % create new cluster
        if(not(matched))
            Pg{i,end+1} = pg;
            P{i,end+1} = p;
        end 
    end
end
 
% run optimization
alignment = groupwiseOrthogonalProcrustes(P, initialAlign);
 
% show found markers in optimized alignment
figure;
hold('on');
for i=1:nSubScans
    % fix Ri to be orthogonal
    [U,~,V] = svd(alignment(i).Rotation);
    Ri = U*V';
    Ti = alignment(i).Translation;
    
    Ea = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);
    
    % bring point cloud into optimized alignment
    pc = pcread(initialAlign(i).FileName);
    tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);
    pcg = pctransform(pc, tform);
   
    pcshow(pcg);
    xlabel('x');
    ylabel('y');
    zlabel('z');
    
    plot3(Ea(:,1), Ea(:,2), Ea(:,3), '.', 'MarkerSize', 15);
    title('Optimized Alignment');
end
 
% write to ALN file
for i=1:length(alignment)
    alignment(i).FileName = initialAlign(i).FileName;
end
 
writeMeshLabALN(alignment, strrep(alnFileName, '.aln', 'Optimized.aln'));
 
end
 
function e = autoDetectMarkers(frame, P, pointCloud)
 
    % create mask based on morphology
    g = rgb2gray(frame);
    % g(g>254) = 0;
    % bw = imbinarize(g, 'adaptive', 'Sensitivity', 10^(-50));
    bw = imbinarize(g, 0.10);
    cc = bwconncomp(bw);
    rp = regionprops(cc, 'Area', 'Solidity', 'Eccentricity', 'Centroid');
    idx = ([rp.Area] > 100 & [rp.Area] < 1000 & [rp.Solidity] > 0.9);
    
    e = cat(1, rp(idx).Centroid);
 
end
 
function e = manuallyDetectMarkers(frame, P, pointCloud)
    
    e = [];
	%edges = edge(rgb2gray(frame), 'Canny', [0.08 0.1], 2);
 
    figure; 
    hold('on');
    imshow(frame);
    title('Close figure to end.');
    set(gcf, 'pointer', 'crosshair'); 
    set(gcf, 'WindowButtonDownFcn', @clickCallback);
    
    uiwait;
 
    function clickCallback(caller, ~)
        
        p = get(gca, 'CurrentPoint'); 
        p = p(1, 1:2);
 
        e = [e; p(:, 1:2)];
        
        if(not(isempty(el)))
            figure(caller);
            hold('on');
            plot(el(1), el(2), 'rx', 'MarkerSize', 15);
        end
    end
    
end
 
function [e, conf] = detectMarkersSubpix(frame, initGuesses, P, Q)
% Detect a marker in a single frame by rectifying to the image and
% performing image registration.
 
    % create mask based on morphology
    g = rgb2gray(frame);
    g(g>254) = 0;
    bw = imbinarize(g);
    cc = bwconncomp(bw);
    labels = labelmatrix(cc);
 
    % project point cloud into image
    q = [Q ones(size(Q,1),1)]*P;
    q = q./[q(:,3) q(:,3) q(:,3)];
    
    e = zeros(size(initGuesses));
    conf = zeros(size(initGuesses, 1), 1);
    
    nMarkersFound = 0;
    
    for i=1:size(initGuesses, 1)
        
        labelId = labels(round(initGuesses(i,2)), round(initGuesses(i,1)));
        labelMask = (labels == labelId);
        labelMask = imdilate(labelMask, strel('disk', 3, 0));
        
        if(sum(sum(labelMask)) < 10 || sum(sum(labelMask)) > 1000)
            continue;
        end
        
        % determine 3D points that are part of the marker
        % note: we should probably undistort labelMask
        pointMask = false(size(q, 1), 1);
        for j=1:size(q,1)
            if(round(q(j,2)) > size(labelMask, 1) || round(q(j,1)) > size(labelMask, 2) || round(q(j,2)) < 1 || round(q(j,1)) < 1)
                continue;
            end
            
            if(labelMask(round(q(j,2)), round(q(j,1))))
                pointMask(j) = true;
            end
        end
        
        if(sum(pointMask)) < 10
            continue;
        end
        
        % project 3D points onto local plane
        [~, sc, ~] = pca(Q(pointMask, :));
        Qlocal = sc(:, 1:2);
        
        % synthetic marker in high res. space
        m = zeros(151, 151);
        [x, y] = meshgrid(1:151, 1:151);
        m((x(:)-76).^2 + (y(:)-76).^2 <= 50^2) = 1.0;
        
        % relation between marker space (px) and true marker/local plane(mm)
        % true marker diameter is 1.75mm
        mScale = 101/1.4; %px/mm
        mShift = 76; %px
        
        % build homography from image to marker space
        H = fitgeotrans(q(pointMask, 1:2), mScale*Qlocal+mShift,  'projective');
        %Hdlt = Hest_DLT([mScale*Qlocal+mShift, ones(size(Qlocal, 1), 1)]', q(pointMask,:)');
        %H = projective2d(Hdlt');
        
        % bring image of marker into marker space
        imMarkerSpace = imwarp(frame, H, 'OutputView', imref2d(size(m)));
        imMarkerSpace = rgb2gray(im2double(imMarkerSpace));
        
        %figure; imshowpair(imMarkerSpace, m);
        
        % perform image registration
        % might be better off using subpixel image correlation
        [opt, met] = imregconfig('multimodal');
        T = imregtform(m, imMarkerSpace, 'translation', opt, met, 'DisplayOptimization', false);
        
        rege = imwarp(m, T, 'OutputView', imref2d(size(m)));
        %figure; imshowpair(imMarkerSpace, rege);
        
        % measure of correlation
        confI = sum(sum(imMarkerSpace .* rege))/sqrt(sum(sum(imMarkerSpace) * sum(sum(rege))));
        %confI = 1.0;
        
        if confI<0.4
            continue;
        end
        
        fprintf('Found marker with confidence: %f\n', confI);
            
        % transform marker space coordinates (76,76) to frame space
        el = T.transformPointsForward([76, 76]);
        el = H.transformPointsInverse(el);
        
        nMarkersFound = nMarkersFound+1;
        e(nMarkersFound,:) = el;
        conf(nMarkersFound) = confI;
    end
    
    e = e(1:nMarkersFound, :);
    conf = conf(1:nMarkersFound);
end
 
function [E, conf] = detectMarkersStereoSubpix(frame0, frame1, initGuesses, camStereoParams, pc)
% Detect a marker in stereo frame set by minimizing the difference to
% projected images of 3d marker
    
    normals = pcnormals(pc, 6);
 
    nMarkers = size(initGuesses, 2);
    
    for i=1:nMarkers
        
        pI = initGuesses(i,:);
    	
        e0 = camStereoParams.CameraParameters1.worldToImage(eye(3,3), zeros(3,1), pI);
        e1 = camStereoParams.CameraParameters2.worldToImage(camStereoParams.RotationOfCamera2, camStereoParams.TranslationOfCamera2, pI);
        
        % center of support region
        e0Center = round(e0);
        e1Center = round(e1);
        
        % find initial normal
        idx = pc.findNearestNeighbors(pI, 1);
        nI = normals(idx, :);
        
        margin = 10;
        
        [x,y] = meshgrid(e0Center(1)-margin:e0Center(1)+margin, e0Center(2)-margin:e0Center(2)+margin);
        e0ImCoords = [x(:), y(:)];
       
        [x,y] = meshgrid(e1Center(1)-margin:e1Center(1)+margin, e1Center(2)-margin:e1Center(2)+margin);
        e1ImCoords = [x(:), y(:)];
        
        e0UndistImCoords = undistortPoints(e0ImCoords, camStereoParams.CameraParameters1);
        e0NormImCoords = camStereoParams.CameraParameters1.pointsToWorld(eye(3,3), zeros(3,1), e0UndistImCoords);
        e1UndistImCoords = undistortPoints(e1ImCoords, camStereoParams.CameraParameters2);
        e1NormImCoords = camStereoParams.CameraParameters2.pointsToWorld(camStereoParams.RotationOfCamera2, camStereoParams.TranslationOfCamera2, e1UndistImCoords);
 
        x0 = [pI; nI(1:2)];
        
        
        
        
    end
    
    function r = circleResiduals(x)
        
        point = x(1:3);
        normal = [x(4:5); sqrt(1-x(4)^2-x(5)^2)];
        
        r = zeros(length(length(e0NormImCoords)) + length(e1NormImCoords), 1);
        
        for i=1:length(e0NormImCoords)
            
            % dermine z coordinate on the hypothesized plane
            z = 
            
        end
        
        
    end
    
end
 
function E = projectOntoPointCloud(e, P, pointCloud)
% Project 2d point coordinates onto pointCloud to find corresponding 3d
% point coordinates.
 
    q = [pointCloud ones(size(pointCloud,1),1)]*P;
    q = q(:,1:2)./[q(:,3) q(:,3)];
 
    E = nan(size(e,1), 3);
    
    for i=1:size(e, 1)
        sqDists = sum((q - repmat(e(i,:), size(q, 1), 1)).^2, 2);
        
        [sqDistsSorted, sortIdx] = sort(sqDists);
        
        neighbors = (sqDistsSorted < 4.0^2);
        
        distsSorted = sqrt(sqDistsSorted(neighbors));
        invDistsSorted = 1.0/distsSorted;
        sortIdx = sortIdx(neighbors);
        
        nNeighbors = sum(neighbors);
        
        if(nNeighbors >= 2)
            E(i, :) = 0;
            for j=1:nNeighbors
                E(i, :) = E(i, :) + invDistsSorted(j)/sum(invDistsSorted) * pointCloud(sortIdx(j), :);
            end
        end
            
    end    
end
 
function H = Hest_DLT(q1, q2)
    % Estimate the homography between a set of point correspondences using the 
    % direct linear transform algorithm.
    %
    % Input:
    %           q1: 3xN matrix of homogenous point coordinates from camera 1. 
    %           q2: 3xN matrix of corresponding points from camera 2.
    % Output:
    %           H: 3x3 matrix. The Fundamental Matrix estimate. 
    %
    % Note that N must be at least 4.
    % See derivation in Aanaes, Lecture Notes on Computer Vision, 2011
 
    % Normalize points
    [T1,invT1] = normalizationMat(q1);
    q1_tilde = T1*q1;
 
    T2 = normalizationMat(q2);
    q2_tilde = T2*q2;
 
    % DLT estimation
    N = size(q1_tilde,2);
    assert(size(q2_tilde,2)==N);
 
    B = zeros(3*N,9);
 
    for i=1:N
        q1i = q1_tilde(:,i);
        q2i = q2_tilde(:,i);
        q1_x = [0 -q1i(3) q1i(2); q1i(3) 0 -q1i(1); -q1i(2) q1i(1) 0];
        biT = kron(q2i', q1_x); 
        B(3*(i-1)+1:3*i, :) = biT;
    end
 
    [U,S,~] = svd(B');
 
    [~,idx] = min(diag(S));
    h = U(:,idx);
 
    H_tilde = reshape(h, 3, 3);
 
    % Unnormalize H
    H = invT1*H_tilde*T2;
 
    % Arbitrarily chose scale
    H = H * 1/H(3,3);
end
 
function [T,invT] = normalizationMat(q)
    % Gives a normalization matrix for homogeneous coordinates
    % such that T*q will have zero mean and unit variance.
    % See Aanaes, Computer Vision Lecture Notes 2.8.2
    %
    % q: (M+1)xN matrix of N MD points in homogenous coordinates
    %
    % Extended to also efficiently compute the inverse matrix
    % DTU, 2013, Jakob Wilm
 
    [M,N] = size(q);
    M = M-1;
 
    mu = mean(q(1:M,:),2);
 
    q_bar = q(1:M,:)-repmat(mu,1,N);
 
    s = mean(sqrt(diag(q_bar'*q_bar)))/sqrt(2);
 
    T = [eye(M)/s, -mu/s; zeros(1,M) 1];
 
    invT = [eye(M)*s, mu; zeros(1,M) 1];
end
 
Generated by GNU Enscript 1.6.6.
 
 
 

Rev 237	Rev 238
1	`function MSE = alignSubScansMarkers(calibrationFileName, alnFileName)`	1	`function MSE = alignSubScansMarkers(calibrationFileName, alnFileName)`
2	`%ALIGNSUBSCANSMARKERS Determines an exact alignment of sub scans (scans`	2	`%ALIGNSUBSCANSMARKERS Determines an exact alignment of sub scans (scans`
3	`% from e.g. one revolution of the rotation stage).`	3	`% from e.g. one revolution of the rotation stage).`
4	`% The method searches for circular white markers of a specific diameter.`	4	`% The method searches for circular white markers of a specific diameter.`
5	`% White frames corresponding to each sub scan must be available.`	5	`% White frames corresponding to each sub scan must be available.`
6	`% A coarse alignment in the form of an aln-file must be provided.`	6	`% A coarse alignment in the form of an aln-file must be provided.`
7	`%`	7	`%`
8	`% 2017 Jakob Wilm, DTU`	8	`% 2017 Jakob Wilm, DTU`
9		9
10	`initialAlign = readMeshLabALN(alnFileName);`	10	`initialAlign = readMeshLabALN(alnFileName);`
11	`[alnFilePath, ~, ~] = fileparts(alnFileName);`	11	`[alnFilePath, ~, ~] = fileparts(alnFileName);`
12		12
13	`calibration = readOpenCVXML(calibrationFileName);`	13	`calibration = readOpenCVXML(calibrationFileName);`
14		14
15	`% correct for Matlab 1-indexing in principle point coordinates`	15	`% correct for Matlab 1-indexing in principle point coordinates`
16	`calibration.K0(1:2, 3) = calibration.K0(1:2, 3)+1;`	16	`calibration.K0(1:2, 3) = calibration.K0(1:2, 3)+1;`
17	`calibration.K1(1:2, 3) = calibration.K1(1:2, 3)+1;`	17	`calibration.K1(1:2, 3) = calibration.K1(1:2, 3)+1;`
18		18
19	`% full projection matrices in Matlab convention`	19	`% full projection matrices in Matlab convention`
20	`P0 = transpose(calibration.K0*[eye(3) zeros(3,1)]);`	20	`P0 = transpose(calibration.K0*[eye(3) zeros(3,1)]);`
21	`P1 = transpose(calibration.K1*[calibration.R1 calibration.T1']);`	21	`P1 = transpose(calibration.K1*[calibration.R1 calibration.T1']);`
22		22
23	`% matlab cam params for undistortion`	23	`% matlab cam params for undistortion`
24	`camParams0 = cameraParameters('IntrinsicMatrix', calibration.K0', 'RadialDistortion', calibration.k0([1 2 5]), 'TangentialDistortion', calibration.k0([3 4]));`	24	`camParams0 = cameraParameters('IntrinsicMatrix', calibration.K0', 'RadialDistortion', calibration.k0([1 2 5]), 'TangentialDistortion', calibration.k0([3 4]));`
25	`camParams1 = cameraParameters('IntrinsicMatrix', calibration.K1', 'RadialDistortion', calibration.k1([1 2 5]), 'TangentialDistortion', calibration.k1([3 4]));`	25	`camParams1 = cameraParameters('IntrinsicMatrix', calibration.K1', 'RadialDistortion', calibration.k1([1 2 5]), 'TangentialDistortion', calibration.k1([3 4]));`
26		26
27	`% matlab struct for triangulation`	27	`% matlab struct for triangulation`
28	`camStereoParams = stereoParameters(camParams0, camParams1, calibration.R1', calibration.T1');`	28	`camStereoParams = stereoParameters(camParams0, camParams1, calibration.R1', calibration.T1');`
29		29
30	`nSubScans = length(initialAlign);`	30	`nSubScans = length(initialAlign);`
31		31
32	`% 3D coordinates of markers in local camera frame`	32	`% 3D coordinates of markers in local camera frame`
33	`E = cell(nSubScans, 1);`	33	`E = cell(nSubScans, 1);`
34		34
35	`% 3D coordinates of markers in global initial alignment`	35	`% 3D coordinates of markers in global initial alignment`
36	`Eg = cell(size(E));`	36	`Eg = cell(size(E));`
37		37
38	`% find 3D markers coordinates`	38	`% find 3D markers coordinates`
39	`for i=1:nSubScans`	39	`for i=1:nSubScans`
40	`%for i=5:5`	40	`%for i=5:5`
41	`% load point cloud`	41	`% load point cloud`
42	`pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);`	42	`pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);`
43	`pcFilePath = fileparts(pcFileName);`	43	`pcFilePath = fileparts(pcFileName);`
44	`pc = pcread(pcFileName);`	44	`pc = pcread(pcFileName);`
45	`Q = pc.Location;`	45	`Q = pc.Location;`
46	`idString = strsplit(initialAlign(i).FileName, {'.ply', '_'});`	46	`idString = strsplit(initialAlign(i).FileName, {'.ply', '_'});`
47	`idString = idString{end-1};`	47	`idString = idString{end-1};`
48		48
49	`% load white frames`	49	`% load white frames`
50	`frame0 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames0_0.png'));`	50	`frame0 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames0_0.png'));`
51	`frame1 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames1_0.png'));`	51	`frame1 = imread(fullfile(pcFilePath, ['sequence_' idString], 'frames1_0.png'));`
52		52
53	`e0Coords = autoDetectMarkers(frame0);`	53	`e0Coords = autoDetectMarkers(frame0);`
54	`e1Coords = autoDetectMarkers(frame1);`	54	`e1Coords = autoDetectMarkers(frame1);`
55		55
56	`%e0Coords = manuallyDetectMarkers(frame0);`	56	`%e0Coords = manuallyDetectMarkers(frame0);`
57	`%e1Coords = manuallyDetectMarkers(frame1);`	57	`%e1Coords = manuallyDetectMarkers(frame1);`
58		58
59	`%[e0Coords, conf0] = detectMarkersSubpix(frame0, e0Coords, P0, Q);`	59	`%[e0Coords, conf0] = detectMarkersSubpix(frame0, e0Coords, P0, Q);`
60	`%[e1Coords, conf1] = detectMarkersSubpix(frame1, e1Coords, P1, Q);`	60	`%[e1Coords, conf1] = detectMarkersSubpix(frame1, e1Coords, P1, Q);`
61		61
62	`if(length(e0Coords) < 1 \|\| length(e1Coords) < 1)`	62	`if(length(e0Coords) < 1 \|\| length(e1Coords) < 1)`
63	`continue;`	63	`continue;`
64	`end`	64	`end`
65		65
66	`% figure;`	66	`% figure;`
67	`% subplot(1,2,1);`	67	`% subplot(1,2,1);`
68	`% imshow(frame0);`	68	`% imshow(frame0);`
69	`% hold('on');`	69	`% hold('on');`
70	`% plot(e0Coords(:,1), e0Coords(:,2), 'rx', 'MarkerSize', 15);`	70	`% plot(e0Coords(:,1), e0Coords(:,2), 'rx', 'MarkerSize', 15);`
71	`% subplot(1,2,2);`	71	`% subplot(1,2,2);`
72	`% imshow(frame1);`	72	`% imshow(frame1);`
73	`% hold('on');`	73	`% hold('on');`
74	`% plot(e1Coords(:,1), e1Coords(:,2), 'rx', 'MarkerSize', 15);`	74	`% plot(e1Coords(:,1), e1Coords(:,2), 'rx', 'MarkerSize', 15);`
75	`% drawnow;`	75	`% drawnow;`
76		76
77	`e0Coords = undistortPoints(e0Coords, camParams0);`	77	`e0Coords = undistortPoints(e0Coords, camParams0);`
78	`e1Coords = undistortPoints(e1Coords, camParams1);`	78	`e1Coords = undistortPoints(e1Coords, camParams1);`
79		79
80	`% match ellipse candidates between cameras based on projection`	80	`% match ellipse candidates between cameras based on projection`
81	`E0 = projectOntoPointCloud(e0Coords, P0, Q);`	81	`E0 = projectOntoPointCloud(e0Coords, P0, Q);`
82	`E1 = projectOntoPointCloud(e1Coords, P1, Q);`	82	`E1 = projectOntoPointCloud(e1Coords, P1, Q);`
83		83
84	`matchedPairs = nan(size(E0, 1), 2);`	84	`matchedPairs = nan(size(E0, 1), 2);`
85	`nMatchedPairs = 0;`	85	`nMatchedPairs = 0;`
86	`for j=1:size(E0, 1)`	86	`for j=1:size(E0, 1)`
87		87
88	`% should use pdist2 instead..`	88	`% should use pdist2 instead..`
89	`sqDists = sum((E1 - repmat(E0(j,:), size(E1, 1), 1)).^2, 2);`	89	`sqDists = sum((E1 - repmat(E0(j,:), size(E1, 1), 1)).^2, 2);`
90		90
91	`[minSqDist, minSqDistIdx] = min(sqDists);`	91	`[minSqDist, minSqDistIdx] = min(sqDists);`
92		92
93	`if(minSqDist < 1^2)`	93	`if(minSqDist < 1^2)`
94	`nMatchedPairs = nMatchedPairs + 1;`	94	`nMatchedPairs = nMatchedPairs + 1;`
95	`matchedPairs(nMatchedPairs, :) = [j, minSqDistIdx];`	95	`matchedPairs(nMatchedPairs, :) = [j, minSqDistIdx];`
96	`end`	96	`end`
97	`end`	97	`end`
98	`matchedPairs = matchedPairs(1:nMatchedPairs, :);`	98	`matchedPairs = matchedPairs(1:nMatchedPairs, :);`
99		99
100	`figure;`	100	`figure;`
101	`subplot(1,2,1);`	101	`subplot(1,2,1);`
102	`imshow(frame0);`	102	`imshow(frame0);`
103	`hold('on');`	103	`hold('on');`
104	`plot(e0Coords(matchedPairs(:, 1),1), e0Coords(matchedPairs(:, 1),2), 'rx', 'MarkerSize', 15);`	104	`plot(e0Coords(matchedPairs(:, 1),1), e0Coords(matchedPairs(:, 1),2), 'rx', 'MarkerSize', 15);`
105	`subplot(1,2,2);`	105	`subplot(1,2,2);`
106	`imshow(frame1);`	106	`imshow(frame1);`
107	`hold('on');`	107	`hold('on');`
108	`plot(e1Coords(matchedPairs(:, 2),1), e1Coords(matchedPairs(:, 2),2), 'rx', 'MarkerSize', 15);`	108	`plot(e1Coords(matchedPairs(:, 2),1), e1Coords(matchedPairs(:, 2),2), 'rx', 'MarkerSize', 15);`
109	`drawnow;`	109	`drawnow;`
110		110
111	`% triangulate marker centers (lens correction has been performed)`	111	`% % triangulate marker centers (lens correction has been performed)`
112	`[E{i}, e] = triangulate(e0Coords(matchedPairs(:, 1),:), e1Coords(matchedPairs(:, 2),:), camStereoParams);`	112	`% [E{i}, e] = triangulate(e0Coords(matchedPairs(:, 1),:), e1Coords(matchedPairs(:, 2),:), camStereoParams);`
113	`E{i} = E{i}(e<3.0, :);`	113	`% E{i} = E{i}(e<3.0, :);`
-		114	`% display(e);`
-		115
-		116	`[E{i}, e] = detectMarkersStereoSubpix(frame0, frame1, E0(matchedPairs(:, 1), :), camStereoParams, pc);`
114	`display(e);`	117	`display(e);`
115		118
116	`% write point cloud with marker (debugging)`	119	`% write point cloud with marker (debugging)`
117	`pcDebug = pointCloud([pc.Location; E{i}], 'Color', [pc.Color; repmat([255, 0, 0], size(E{i}, 1), 1)]);`	120	`pcDebug = pointCloud([pc.Location; E{i}], 'Color', [pc.Color; repmat([255, 0, 0], size(E{i}, 1), 1)]);`
118	`pcwrite(pcDebug, 'pcDebug.ply');`	121	`pcwrite(pcDebug, 'pcDebug.ply');`
119		122
120	`% bring markers into initial alignment`	123	`% bring markers into initial alignment`
121	`[U,~,V] = svd(initialAlign(i).Rotation);`	124	`[U,~,V] = svd(initialAlign(i).Rotation);`
122	`Ri = U*V';`	125	`Ri = U*V';`
123	`Ti = initialAlign(i).Translation;`	126	`Ti = initialAlign(i).Translation;`
124		127
125	`Eg{i} = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);`	128	`Eg{i} = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);`
126	`end`	129	`end`
127		130
128	`% show found markers in initial alignment`	131	`% show found markers in initial alignment`
129	`figure;`	132	`figure;`
130	`hold('on');`	133	`hold('on');`
131	`for i=1:nSubScans`	134	`for i=1:nSubScans`
132	`% fix Ri to be orthogonal`	135	`% fix Ri to be orthogonal`
133	`[U,~,V] = svd(initialAlign(i).Rotation);`	136	`[U,~,V] = svd(initialAlign(i).Rotation);`
134	`Ri = U*V';`	137	`Ri = U*V';`
135		138
136	`% bring point cloud into initial alignment`	139	`% bring point cloud into initial alignment`
137	`pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);`	140	`pcFileName = fullfile(alnFilePath, initialAlign(i).FileName);`
138	`pc = pcread(pcFileName);`	141	`pc = pcread(pcFileName);`
139	`tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);`	142	`tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);`
140	`pcg = pctransform(pc, tform);`	143	`pcg = pctransform(pc, tform);`
141		144
142	`pcshow(pcg);`	145	`pcshow(pcg);`
143	`xlabel('x');`	146	`xlabel('x');`
144	`ylabel('y');`	147	`ylabel('y');`
145	`zlabel('z');`	148	`zlabel('z');`
146		149
147	`plot3(Eg{i}(:,1), Eg{i}(:,2), Eg{i}(:,3), '.', 'MarkerSize', 15);`	150	`plot3(Eg{i}(:,1), Eg{i}(:,2), Eg{i}(:,3), '.', 'MarkerSize', 15);`
148	`title('Initial Alignment');`	151	`title('Initial Alignment');`
149	`end`	152	`end`
150		153
151	`% match markers between poses using initial alignment`	154	`% match markers between poses using initial alignment`
152	`Pg = {};`	155	`Pg = {};`
153	`P = {};`	156	`P = {};`
154	`for i=1:nSubScans`	157	`for i=1:nSubScans`
155	`for j=1:size(Eg{i}, 1)`	158	`for j=1:size(Eg{i}, 1)`
156	`pg = Eg{i}(j,:);`	159	`pg = Eg{i}(j,:);`
157	`p = E{i}(j,:);`	160	`p = E{i}(j,:);`
158	`matched = false;`	161	`matched = false;`
159	`for k=1:size(Pg, 2)`	162	`for k=1:size(Pg, 2)`
160	`clusterCenter = mean(cat(1, Pg{:,k}), 1);`	163	`clusterCenter = mean(cat(1, Pg{:,k}), 1);`
161	`if(sum((pg - clusterCenter).^2) < 3^2)`	164	`if(sum((pg - clusterCenter).^2) < 3^2)`
162	`% store in global frame`	165	`% store in global frame`
163	`Pg{i,k} = pg;`	166	`Pg{i,k} = pg;`
164	`% store in local frame`	167	`% store in local frame`
165	`P{i,k} = p;`	168	`P{i,k} = p;`
166	`matched = true;`	169	`matched = true;`
167	`break;`	170	`break;`
168	`end`	171	`end`
169	`end`	172	`end`
170	`% create new cluster`	173	`% create new cluster`
171	`if(not(matched))`	174	`if(not(matched))`
172	`Pg{i,end+1} = pg;`	175	`Pg{i,end+1} = pg;`
173	`P{i,end+1} = p;`	176	`P{i,end+1} = p;`
174	`end`	177	`end`
175	`end`	178	`end`
176	`end`	179	`end`
177		180
178	`% run optimization`	181	`% run optimization`
179	`alignment = groupwiseOrthogonalProcrustes(P, initialAlign);`	182	`alignment = groupwiseOrthogonalProcrustes(P, initialAlign);`
180		183
181	`% show found markers in optimized alignment`	184	`% show found markers in optimized alignment`
182	`figure;`	185	`figure;`
183	`hold('on');`	186	`hold('on');`
184	`for i=1:nSubScans`	187	`for i=1:nSubScans`
185	`% fix Ri to be orthogonal`	188	`% fix Ri to be orthogonal`
186	`[U,~,V] = svd(alignment(i).Rotation);`	189	`[U,~,V] = svd(alignment(i).Rotation);`
187	`Ri = U*V';`	190	`Ri = U*V';`
188	`Ti = alignment(i).Translation;`	191	`Ti = alignment(i).Translation;`
189		192
190	`Ea = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);`	193	`Ea = E{i}*Ri' + repmat(Ti', size(E{i}, 1), 1);`
191		194
192	`% bring point cloud into optimized alignment`	195	`% bring point cloud into optimized alignment`
193	`pc = pcread(initialAlign(i).FileName);`	196	`pc = pcread(initialAlign(i).FileName);`
194	`tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);`	197	`tform = affine3d([Ri' [0;0;0]; initialAlign(i).Translation' 1]);`
195	`pcg = pctransform(pc, tform);`	198	`pcg = pctransform(pc, tform);`
196		199
197	`pcshow(pcg);`	200	`pcshow(pcg);`
198	`xlabel('x');`	201	`xlabel('x');`
199	`ylabel('y');`	202	`ylabel('y');`
200	`zlabel('z');`	203	`zlabel('z');`
201		204
202	`plot3(Ea(:,1), Ea(:,2), Ea(:,3), '.', 'MarkerSize', 15);`	205	`plot3(Ea(:,1), Ea(:,2), Ea(:,3), '.', 'MarkerSize', 15);`
203	`title('Optimized Alignment');`	206	`title('Optimized Alignment');`
204	`end`	207	`end`
205		208
206	`% write to ALN file`	209	`% write to ALN file`
207	`for i=1:length(alignment)`	210	`for i=1:length(alignment)`
208	`alignment(i).FileName = initialAlign(i).FileName;`	211	`alignment(i).FileName = initialAlign(i).FileName;`
209	`end`	212	`end`
210		213
211	`writeMeshLabALN(alignment, strrep(alnFileName, '.aln', 'Optimized.aln'));`	214	`writeMeshLabALN(alignment, strrep(alnFileName, '.aln', 'Optimized.aln'));`
212		215
213	`end`	216	`end`
214		217
215	`function e = autoDetectMarkers(frame, P, pointCloud)`	218	`function e = autoDetectMarkers(frame, P, pointCloud)`
216		219
217	`% create mask based on morphology`	220	`% create mask based on morphology`
218	`g = rgb2gray(frame);`	221	`g = rgb2gray(frame);`
219	`% g(g>254) = 0;`	222	`% g(g>254) = 0;`
220	`% bw = imbinarize(g, 'adaptive', 'Sensitivity', 10^(-50));`	223	`% bw = imbinarize(g, 'adaptive', 'Sensitivity', 10^(-50));`
221	`bw = imbinarize(g, 0.10);`	224	`bw = imbinarize(g, 0.10);`
222	`cc = bwconncomp(bw);`	225	`cc = bwconncomp(bw);`
223	`rp = regionprops(cc, 'Area', 'Solidity', 'Eccentricity', 'Centroid');`	226	`rp = regionprops(cc, 'Area', 'Solidity', 'Eccentricity', 'Centroid');`
224	`idx = ([rp.Area] > 100 & [rp.Area] < 1000 & [rp.Solidity] > 0.9);`	227	`idx = ([rp.Area] > 100 & [rp.Area] < 1000 & [rp.Solidity] > 0.9);`
225		228
226	`e = cat(1, rp(idx).Centroid);`	229	`e = cat(1, rp(idx).Centroid);`
227		230
228	`end`	231	`end`
229		232
230	`function e = manuallyDetectMarkers(frame, P, pointCloud)`	233	`function e = manuallyDetectMarkers(frame, P, pointCloud)`
231		234
232	`e = [];`	235	`e = [];`
233	`%edges = edge(rgb2gray(frame), 'Canny', [0.08 0.1], 2);`	236	`%edges = edge(rgb2gray(frame), 'Canny', [0.08 0.1], 2);`
234		237
235	`figure;`	238	`figure;`
236	`hold('on');`	239	`hold('on');`
237	`imshow(frame);`	240	`imshow(frame);`
238	`title('Close figure to end.');`	241	`title('Close figure to end.');`
239	`set(gcf, 'pointer', 'crosshair');`	242	`set(gcf, 'pointer', 'crosshair');`
240	`set(gcf, 'WindowButtonDownFcn', @clickCallback);`	243	`set(gcf, 'WindowButtonDownFcn', @clickCallback);`
241		244
242	`uiwait;`	245	`uiwait;`
243		246
244	`function clickCallback(caller, ~)`	247	`function clickCallback(caller, ~)`
245		248
246	`p = get(gca, 'CurrentPoint');`	249	`p = get(gca, 'CurrentPoint');`
247	`p = p(1, 1:2);`	250	`p = p(1, 1:2);`
248		251
249	`e = [e; p(:, 1:2)];`	252	`e = [e; p(:, 1:2)];`
250		253
251	`if(not(isempty(el)))`	254	`if(not(isempty(el)))`
252	`figure(caller);`	255	`figure(caller);`
253	`hold('on');`	256	`hold('on');`
254	`plot(el(1), el(2), 'rx', 'MarkerSize', 15);`	257	`plot(el(1), el(2), 'rx', 'MarkerSize', 15);`
255	`end`	258	`end`
256	`end`	259	`end`
257		260
258	`end`	261	`end`
259		262
260	`function [e, conf] = detectMarkersSubpix(frame, initGuesses, P, Q)`	263	`function [e, conf] = detectMarkersSubpix(frame, initGuesses, P, Q)`
261	`% Detect a marker in a single frame by rectifying to the image and`	264	`% Detect a marker in a single frame by rectifying to the image and`
262	`% performing image registration.`	265	`% performing image registration.`
263		266
264	`% create mask based on morphology`	267	`% create mask based on morphology`
265	`g = rgb2gray(frame);`	268	`g = rgb2gray(frame);`
266	`g(g>254) = 0;`	269	`g(g>254) = 0;`
267	`bw = imbinarize(g);`	270	`bw = imbinarize(g);`
268	`cc = bwconncomp(bw);`	271	`cc = bwconncomp(bw);`
269	`labels = labelmatrix(cc);`	272	`labels = labelmatrix(cc);`
270		273
271	`% project point cloud into image`	274	`% project point cloud into image`
272	`q = [Q ones(size(Q,1),1)]*P;`	275	`q = [Q ones(size(Q,1),1)]*P;`
273	`q = q./[q(:,3) q(:,3) q(:,3)];`	276	`q = q./[q(:,3) q(:,3) q(:,3)];`
274		277
275	`e = zeros(size(initGuesses));`	278	`e = zeros(size(initGuesses));`
276	`conf = zeros(size(initGuesses, 1), 1);`	279	`conf = zeros(size(initGuesses, 1), 1);`
277		280
278	`nMarkersFound = 0;`	281	`nMarkersFound = 0;`
279		282
280	`for i=1:size(initGuesses, 1)`	283	`for i=1:size(initGuesses, 1)`
281		284
282	`labelId = labels(round(initGuesses(i,2)), round(initGuesses(i,1)));`	285	`labelId = labels(round(initGuesses(i,2)), round(initGuesses(i,1)));`
283	`labelMask = (labels == labelId);`	286	`labelMask = (labels == labelId);`
284	`labelMask = imdilate(labelMask, strel('disk', 3, 0));`	287	`labelMask = imdilate(labelMask, strel('disk', 3, 0));`
285		288
286	`if(sum(sum(labelMask)) < 10 \|\| sum(sum(labelMask)) > 1000)`	289	`if(sum(sum(labelMask)) < 10 \|\| sum(sum(labelMask)) > 1000)`
287	`continue;`	290	`continue;`
288	`end`	291	`end`
289		292
290	`% determine 3D points that are part of the marker`	293	`% determine 3D points that are part of the marker`
291	`% note: we should probably undistort labelMask`	294	`% note: we should probably undistort labelMask`
292	`pointMask = false(size(q, 1), 1);`	295	`pointMask = false(size(q, 1), 1);`
293	`for j=1:size(q,1)`	296	`for j=1:size(q,1)`
294	`if(round(q(j,2)) > size(labelMask, 1) \|\| round(q(j,1)) > size(labelMask, 2) \|\| round(q(j,2)) < 1 \|\| round(q(j,1)) < 1)`	297	`if(round(q(j,2)) > size(labelMask, 1) \|\| round(q(j,1)) > size(labelMask, 2) \|\| round(q(j,2)) < 1 \|\| round(q(j,1)) < 1)`
295	`continue;`	298	`continue;`
296	`end`	299	`end`
297		300
298	`if(labelMask(round(q(j,2)), round(q(j,1))))`	301	`if(labelMask(round(q(j,2)), round(q(j,1))))`
299	`pointMask(j) = true;`	302	`pointMask(j) = true;`
300	`end`	303	`end`
301	`end`	304	`end`
302		305
303	`if(sum(pointMask)) < 10`	306	`if(sum(pointMask)) < 10`
304	`continue;`	307	`continue;`
305	`end`	308	`end`
306		309
307	`% project 3D points onto local plane`	310	`% project 3D points onto local plane`
308	`[~, sc, ~] = pca(Q(pointMask, :));`	311	`[~, sc, ~] = pca(Q(pointMask, :));`
309	`Qlocal = sc(:, 1:2);`	312	`Qlocal = sc(:, 1:2);`
310		313
311	`% synthetic marker in high res. space`	314	`% synthetic marker in high res. space`
312	`m = zeros(151, 151);`	315	`m = zeros(151, 151);`
313	`[x, y] = meshgrid(1:151, 1:151);`	316	`[x, y] = meshgrid(1:151, 1:151);`
314	`m((x(:)-76).^2 + (y(:)-76).^2 <= 50^2) = 1.0;`	317	`m((x(:)-76).^2 + (y(:)-76).^2 <= 50^2) = 1.0;`
315		318
316	`% relation between marker space (px) and true marker/local plane(mm)`	319	`% relation between marker space (px) and true marker/local plane(mm)`
317	`% true marker diameter is 1.75mm`	320	`% true marker diameter is 1.75mm`
318	`mScale = 101/1.4; %px/mm`	321	`mScale = 101/1.4; %px/mm`
319	`mShift = 76; %px`	322	`mShift = 76; %px`
320		323
321	`% build homography from image to marker space`	324	`% build homography from image to marker space`
322	`H = fitgeotrans(q(pointMask, 1:2), mScale*Qlocal+mShift, 'projective');`	325	`H = fitgeotrans(q(pointMask, 1:2), mScale*Qlocal+mShift, 'projective');`
323	`%Hdlt = Hest_DLT([mScale*Qlocal+mShift, ones(size(Qlocal, 1), 1)]', q(pointMask,:)');`	326	`%Hdlt = Hest_DLT([mScale*Qlocal+mShift, ones(size(Qlocal, 1), 1)]', q(pointMask,:)');`
324	`%H = projective2d(Hdlt');`	327	`%H = projective2d(Hdlt');`
325		328
326	`% bring image of marker into marker space`	329	`% bring image of marker into marker space`
327	`imMarkerSpace = imwarp(frame, H, 'OutputView', imref2d(size(m)));`	330	`imMarkerSpace = imwarp(frame, H, 'OutputView', imref2d(size(m)));`
328	`imMarkerSpace = rgb2gray(im2double(imMarkerSpace));`	331	`imMarkerSpace = rgb2gray(im2double(imMarkerSpace));`
329		332
330	`%figure; imshowpair(imMarkerSpace, m);`	333	`%figure; imshowpair(imMarkerSpace, m);`
331		334
332	`% perform image registration`	335	`% perform image registration`
333	`% might be better off using subpixel image correlation`	336	`% might be better off using subpixel image correlation`
334	`[opt, met] = imregconfig('multimodal');`	337	`[opt, met] = imregconfig('multimodal');`
335	`T = imregtform(m, imMarkerSpace, 'translation', opt, met, 'DisplayOptimization', false);`	338	`T = imregtform(m, imMarkerSpace, 'translation', opt, met, 'DisplayOptimization', false);`
336		339
337	`rege = imwarp(m, T, 'OutputView', imref2d(size(m)));`	340	`rege = imwarp(m, T, 'OutputView', imref2d(size(m)));`
338	`%figure; imshowpair(imMarkerSpace, rege);`	341	`%figure; imshowpair(imMarkerSpace, rege);`
339		342
340	`% measure of correlation`	343	`% measure of correlation`
341	`confI = sum(sum(imMarkerSpace .* rege))/sqrt(sum(sum(imMarkerSpace) * sum(sum(rege))));`	344	`confI = sum(sum(imMarkerSpace .* rege))/sqrt(sum(sum(imMarkerSpace) * sum(sum(rege))));`
342	`%confI = 1.0;`	345	`%confI = 1.0;`
343		346
344	`if confI<0.4`	347	`if confI<0.4`
345	`continue;`	348	`continue;`
346	`end`	349	`end`
347		350
348	`fprintf('Found marker with confidence: %f\n', confI);`	351	`fprintf('Found marker with confidence: %f\n', confI);`
349		352
350	`% transform marker space coordinates (76,76) to frame space`	353	`% transform marker space coordinates (76,76) to frame space`
351	`el = T.transformPointsForward([76, 76]);`	354	`el = T.transformPointsForward([76, 76]);`
352	`el = H.transformPointsInverse(el);`	355	`el = H.transformPointsInverse(el);`
353		356
354	`nMarkersFound = nMarkersFound+1;`	357	`nMarkersFound = nMarkersFound+1;`
355	`e(nMarkersFound,:) = el;`	358	`e(nMarkersFound,:) = el;`
356	`conf(nMarkersFound) = confI;`	359	`conf(nMarkersFound) = confI;`
357	`end`	360	`end`
358		361
359	`e = e(1:nMarkersFound, :);`	362	`e = e(1:nMarkersFound, :);`
360	`conf = conf(1:nMarkersFound);`	363	`conf = conf(1:nMarkersFound);`
361	`end`	364	`end`
-		365
-		366	`function [E, conf] = detectMarkersStereoSubpix(frame0, frame1, initGuesses, camStereoParams, pc)`
-		367	`% Detect a marker in stereo frame set by minimizing the difference to`
-		368	`% projected images of 3d marker`
-		369
-		370	`normals = pcnormals(pc, 6);`
-		371
-		372	`nMarkers = size(initGuesses, 2);`
-		373
-		374	`for i=1:nMarkers`
-		375
-		376	`pI = initGuesses(i,:);`
-		377
-		378	`e0 = camStereoParams.CameraParameters1.worldToImage(eye(3,3), zeros(3,1), pI);`
-		379	`e1 = camStereoParams.CameraParameters2.worldToImage(camStereoParams.RotationOfCamera2, camStereoParams.TranslationOfCamera2, pI);`
-		380
-		381	`% center of support region`
-		382	`e0Center = round(e0);`
-		383	`e1Center = round(e1);`
-		384
-		385	`% find initial normal`
-		386	`idx = pc.findNearestNeighbors(pI, 1);`
-		387	`nI = normals(idx, :);`
-		388
-		389	`margin = 10;`
-		390
-		391	`[x,y] = meshgrid(e0Center(1)-margin:e0Center(1)+margin, e0Center(2)-margin:e0Center(2)+margin);`
-		392	`e0ImCoords = [x(:), y(:)];`
-		393
-		394	`[x,y] = meshgrid(e1Center(1)-margin:e1Center(1)+margin, e1Center(2)-margin:e1Center(2)+margin);`
-		395	`e1ImCoords = [x(:), y(:)];`
-		396
-		397	`e0UndistImCoords = undistortPoints(e0ImCoords, camStereoParams.CameraParameters1);`
-		398	`e0NormImCoords = camStereoParams.CameraParameters1.pointsToWorld(eye(3,3), zeros(3,1), e0UndistImCoords);`
-		399	`e1UndistImCoords = undistortPoints(e1ImCoords, camStereoParams.CameraParameters2);`
-		400	`e1NormImCoords = camStereoParams.CameraParameters2.pointsToWorld(camStereoParams.RotationOfCamera2, camStereoParams.TranslationOfCamera2, e1UndistImCoords);`
-		401
-		402	`x0 = [pI; nI(1:2)];`
-		403
-		404
-		405
-		406
-		407	`end`
-		408
-		409	`function r = circleResiduals(x)`
-		410
-		411	`point = x(1:3);`
-		412	`normal = [x(4:5); sqrt(1-x(4)^2-x(5)^2)];`
-		413
-		414	`r = zeros(length(length(e0NormImCoords)) + length(e1NormImCoords), 1);`
-		415
-		416	`for i=1:length(e0NormImCoords)`
-		417
-		418	`% dermine z coordinate on the hypothesized plane`
-		419	`z =`
-		420
-		421	`end`
-		422
-		423
-		424	`end`
-		425
-		426	`end`
362		427
363	`function E = projectOntoPointCloud(e, P, pointCloud)`	428	`function E = projectOntoPointCloud(e, P, pointCloud)`
364	`% Project 2d point coordinates onto pointCloud to find corresponding 3d`	429	`% Project 2d point coordinates onto pointCloud to find corresponding 3d`
365	`% point coordinates.`	430	`% point coordinates.`
366		431
367	`q = [pointCloud ones(size(pointCloud,1),1)]*P;`	432	`q = [pointCloud ones(size(pointCloud,1),1)]*P;`
368	`q = q(:,1:2)./[q(:,3) q(:,3)];`	433	`q = q(:,1:2)./[q(:,3) q(:,3)];`
369		434
370	`E = nan(size(e,1), 3);`	435	`E = nan(size(e,1), 3);`
371		436
372	`for i=1:size(e, 1)`	437	`for i=1:size(e, 1)`
373	`sqDists = sum((q - repmat(e(i,:), size(q, 1), 1)).^2, 2);`	438	`sqDists = sum((q - repmat(e(i,:), size(q, 1), 1)).^2, 2);`
374		439
375	`[sqDistsSorted, sortIdx] = sort(sqDists);`	440	`[sqDistsSorted, sortIdx] = sort(sqDists);`
376		441
377	`neighbors = (sqDistsSorted < 4.0^2);`	442	`neighbors = (sqDistsSorted < 4.0^2);`
378		443
379	`distsSorted = sqrt(sqDistsSorted(neighbors));`	444	`distsSorted = sqrt(sqDistsSorted(neighbors));`
380	`invDistsSorted = 1.0/distsSorted;`	445	`invDistsSorted = 1.0/distsSorted;`
381	`sortIdx = sortIdx(neighbors);`	446	`sortIdx = sortIdx(neighbors);`
382		447
383	`nNeighbors = sum(neighbors);`	448	`nNeighbors = sum(neighbors);`
384		449
385	`if(nNeighbors >= 2)`	450	`if(nNeighbors >= 2)`
386	`E(i, :) = 0;`	451	`E(i, :) = 0;`
387	`for j=1:nNeighbors`	452	`for j=1:nNeighbors`
388	`E(i, :) = E(i, :) + invDistsSorted(j)/sum(invDistsSorted) * pointCloud(sortIdx(j), :);`	453	`E(i, :) = E(i, :) + invDistsSorted(j)/sum(invDistsSorted) * pointCloud(sortIdx(j), :);`
389	`end`	454	`end`
390	`end`	455	`end`
391		456
392	`end`	457	`end`
393	`end`	458	`end`
394		459
395	`function H = Hest_DLT(q1, q2)`	460	`function H = Hest_DLT(q1, q2)`
396	`% Estimate the homography between a set of point correspondences using the`	461	`% Estimate the homography between a set of point correspondences using the`
397	`% direct linear transform algorithm.`	462	`% direct linear transform algorithm.`
398	`%`	463	`%`
399	`% Input:`	464	`% Input:`
400	`% q1: 3xN matrix of homogenous point coordinates from camera 1.`	465	`% q1: 3xN matrix of homogenous point coordinates from camera 1.`
401	`% q2: 3xN matrix of corresponding points from camera 2.`	466	`% q2: 3xN matrix of corresponding points from camera 2.`
402	`% Output:`	467	`% Output:`
403	`% H: 3x3 matrix. The Fundamental Matrix estimate.`	468	`% H: 3x3 matrix. The Fundamental Matrix estimate.`
404	`%`	469	`%`
405	`% Note that N must be at least 4.`	470	`% Note that N must be at least 4.`
406	`% See derivation in Aanaes, Lecture Notes on Computer Vision, 2011`	471	`% See derivation in Aanaes, Lecture Notes on Computer Vision, 2011`
407		472
408	`% Normalize points`	473	`% Normalize points`
409	`[T1,invT1] = normalizationMat(q1);`	474	`[T1,invT1] = normalizationMat(q1);`
410	`q1_tilde = T1*q1;`	475	`q1_tilde = T1*q1;`
411		476
412	`T2 = normalizationMat(q2);`	477	`T2 = normalizationMat(q2);`
413	`q2_tilde = T2*q2;`	478	`q2_tilde = T2*q2;`
414		479
415	`% DLT estimation`	480	`% DLT estimation`
416	`N = size(q1_tilde,2);`	481	`N = size(q1_tilde,2);`
417	`assert(size(q2_tilde,2)==N);`	482	`assert(size(q2_tilde,2)==N);`
418		483
419	`B = zeros(3*N,9);`	484	`B = zeros(3*N,9);`
420		485
421	`for i=1:N`	486	`for i=1:N`
422	`q1i = q1_tilde(:,i);`	487	`q1i = q1_tilde(:,i);`
423	`q2i = q2_tilde(:,i);`	488	`q2i = q2_tilde(:,i);`
424	`q1_x = [0 -q1i(3) q1i(2); q1i(3) 0 -q1i(1); -q1i(2) q1i(1) 0];`	489	`q1_x = [0 -q1i(3) q1i(2); q1i(3) 0 -q1i(1); -q1i(2) q1i(1) 0];`
425	`biT = kron(q2i', q1_x);`	490	`biT = kron(q2i', q1_x);`
426	`B(3(i-1)+1:3i, :) = biT;`	491	`B(3(i-1)+1:3i, :) = biT;`
427	`end`	492	`end`
428		493
429	`[U,S,~] = svd(B');`	494	`[U,S,~] = svd(B');`
430		495
431	`[~,idx] = min(diag(S));`	496	`[~,idx] = min(diag(S));`
432	`h = U(:,idx);`	497	`h = U(:,idx);`
433		498
434	`H_tilde = reshape(h, 3, 3);`	499	`H_tilde = reshape(h, 3, 3);`
435		500

Subversion Repositories seema-scanner

(root)/matlab/alignSubScansMarkers.m – Rev 237 → 238