在MATLAB中使用pointCloud时,“颜色”必须与输入点数相对应

时间:2019-02-12 14:31:07

标签: matlab camera-calibration point-clouds disparity-mapping

我想用立体摄像机创建3D地图,为了测试这一点,我使用了两个给定的MATLAB示例:

  1. https://de.mathworks.com/help/vision/ref/estimatecameraparameters.html (“立体声相机校准”)
  2. https://de.mathworks.com/help/vision/examples/depth-estimation-from-stereo-video.html

我将这两个脚本合并为以下脚本:

% load left and right images
leftImages = imageDatastore(fullfile(toolboxdir('vision'),'visiondata', ...
    'calibration','stereo','left'));
rightImages = imageDatastore(fullfile(toolboxdir('vision'),'visiondata', ...
    'calibration','stereo','right'));

% calculate image points
[imagePoints,boardSize] = ...
  detectCheckerboardPoints(leftImages.Files,rightImages.Files);

% calculate world points
squareSize = 108;
worldPoints = generateCheckerboardPoints(boardSize,squareSize);


% calculate camera paramters
I = readimage(leftImages,1); 
imageSize = [size(I,1),size(I,2)];
stereoParams = estimateCameraParameters(imagePoints,worldPoints, ...
                                  'ImageSize',imageSize);


% get left and right image
frameLeftGray = imread(leftImages.Files{1});
frameRightGray = imread(rightImages.Files{1});
[frameLeftRect, frameRightRect] = ...
    rectifyStereoImages(frameLeftGray, frameRightGray, stereoParams);


% get disparity map    
disparityMap = disparity(frameLeftRect, frameRightRect);
figure;
imshow(disparityMap, [0, 128]);
title('Disparity Map');
colormap jet
colorbar

% create 3D Bar 
points3D = reconstructScene(disparityMap, stereoParams);

% Convert to meters and create a pointCloud object
points3D = points3D ./ 1000;

% This will fail
ptCloud = pointCloud(points3D, 'Color', frameLeftRect);

% Create a streaming point cloud viewer
player3D = pcplayer([-3, 3], [-3, 3], [0, 8], 'VerticalAxis', 'y', ...
    'VerticalAxisDir', 'down');

% Visualize the point cloud
view(player3D, ptCloud);

但是,执行错误的我会收到以下错误消息:

  

使用pointCloud / set.Color(第545行)时出现错误,“ Color”必须与   输入点数。

     

pointCloud中的错误(第151行)               this.Color = C;

     

DepthEstimation中的错误(第45行)ptCloud = pointCloud(points3D,   'Color',frameLeftRect);

分别尝试示例1)和2)时,它们工作得很好。我认为这与图像尺寸本身有关。但是,调整大小会导致相机参数错误。

那么还有其他方法可以解决存在“颜色”参数的错误?

提前谢谢

1 个答案:

答案 0 :(得分:1)

您使用的是灰度图像作为输入,因此它与RGB点不匹配。从灰度图像创建RGB图像,然后使用它。

rgb = cat(3,frameRightRect,frameRightRect,frameRightRect);
ptCloud = pointCloud(points3D, 'Color', rgb);