Bringing the RGB color back to image after processing

Question

Ghulam 2018-2-27

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/385060-bringing-the-rgb-color-back-to-image-after-processing

回答： Ghulam 2018-2-27

Hi, I am performing some type of processing on images due to which they are alligned with each other and are shifted in circular direction.

I want to bring the origional colors back to the images after the processing is completed. I have tried imposing the channel colors back but that didnt worked. It returned the images to its color and shape too.

Can anyone please guide me how can i achieve it?

Thanks a lot.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

KSSV 2018-2-27

Show us your code..so that we can help.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Ghulam 2018-2-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/385060-bringing-the-rgb-color-back-to-image-after-processing#answer_307322

在 MATLAB Online 中打开

Here is the code

if true
  % code
FIXED = imread('a1.jpg');
MOVING1 = imread('a2.jpg');
MOVING2 = imread('a3.jpg');
MOVING3 = imread('a4.jpg');
MOVING4 = imread('a5.jpg');

% Convert RGB images to grayscale FIXED = rgb2gray(FIXED); MOVING1 = rgb2gray(MOVING1); MOVING2 = rgb2gray(MOVING2); MOVING3 = rgb2gray(MOVING3); MOVING4 = rgb2gray(MOVING4);

% Default spatial referencing objects fixedRefObj = imref2d(size(FIXED)); movingRefObj1 = imref2d(size(MOVING1)); movingRefObj2 = imref2d(size(MOVING2)); movingRefObj3 = imref2d(size(MOVING3)); movingRefObj4 = imref2d(size(MOVING4));

% Detect SURF features fixedPoints = detectSURFFeatures(FIXED,'MetricThreshold',750.000000,'NumOctaves',3,'NumScaleLevels',5); movingPoints1 = detectSURFFeatures(MOVING1,'MetricThreshold',750.000000,'NumOctaves',3,'NumScaleLevels',5); movingPoints2 = detectSURFFeatures(MOVING2,'MetricThreshold',750.000000,'NumOctaves',3,'NumScaleLevels',5); movingPoints3 = detectSURFFeatures(MOVING3,'MetricThreshold',750.000000,'NumOctaves',3,'NumScaleLevels',5); movingPoints4 = detectSURFFeatures(MOVING4,'MetricThreshold',750.000000,'NumOctaves',3,'NumScaleLevels',5);

% Extract features [fixedFeatures,fixedValidPoints] = extractFeatures(FIXED,fixedPoints,'Upright',false); [movingFeatures1,movingValidPoints1] = extractFeatures(MOVING1,movingPoints1,'Upright',false); [movingFeatures2,movingValidPoints2] = extractFeatures(MOVING2,movingPoints2,'Upright',false); [movingFeatures3,movingValidPoints3] = extractFeatures(MOVING3,movingPoints3,'Upright',false); [movingFeatures4,movingValidPoints4] = extractFeatures(MOVING4,movingPoints4,'Upright',false);

% Match features indexPairs1 = matchFeatures(fixedFeatures,movingFeatures1,'MatchThreshold',42.708333,'MaxRatio',0.427083); indexPairs2 = matchFeatures(fixedFeatures,movingFeatures2,'MatchThreshold',42.708333,'MaxRatio',0.427083); indexPairs3 = matchFeatures(fixedFeatures,movingFeatures3,'MatchThreshold',42.708333,'MaxRatio',0.427083); indexPairs4 = matchFeatures(fixedFeatures,movingFeatures4,'MatchThreshold',42.708333,'MaxRatio',0.427083);

fixedMatchedPoints1 = fixedValidPoints(indexPairs1(:,1)); fixedMatchedPoints2 = fixedValidPoints(indexPairs2(:,1)); fixedMatchedPoints3 = fixedValidPoints(indexPairs3(:,1)); fixedMatchedPoints4 = fixedValidPoints(indexPairs4(:,1));

movingMatchedPoints1 = movingValidPoints1(indexPairs1(:,2)); movingMatchedPoints2 = movingValidPoints2(indexPairs2(:,2)); movingMatchedPoints3 = movingValidPoints3(indexPairs3(:,2)); movingMatchedPoints4 = movingValidPoints4(indexPairs4(:,2));

MOVINGREG.FixedMatchedFeatures1 = fixedMatchedPoints1; MOVINGREG.FixedMatchedFeatures2 = fixedMatchedPoints2; MOVINGREG.FixedMatchedFeatures3 = fixedMatchedPoints3; MOVINGREG.FixedMatchedFeatures4 = fixedMatchedPoints4;

MOVINGREG.MovingMatchedFeatures1 = movingMatchedPoints1; MOVINGREG.MovingMatchedFeatures2 = movingMatchedPoints2; MOVINGREG.MovingMatchedFeatures3 = movingMatchedPoints3; MOVINGREG.MovingMatchedFeatures4 = movingMatchedPoints4;

% Apply transformation - Results may not be identical between runs because of the randomized nature of the algorithm tform1 = estimateGeometricTransform(movingMatchedPoints1,fixedMatchedPoints1,'affine'); tform2 = estimateGeometricTransform(movingMatchedPoints2,fixedMatchedPoints2,'affine'); tform3 = estimateGeometricTransform(movingMatchedPoints3,fixedMatchedPoints3,'affine'); tform4 = estimateGeometricTransform(movingMatchedPoints4,fixedMatchedPoints4,'affine');

MOVINGREG.Transformation1 = tform1; MOVINGREG.Transformation2 = tform2; MOVINGREG.Transformation3 = tform3; MOVINGREG.Transformation4 = tform4;

MOVINGREG.RegisteredImage1 = imwarp(MOVING1, movingRefObj1, tform1, 'OutputView', fixedRefObj, 'SmoothEdges', true); MOVINGREG.RegisteredImage2 = imwarp(MOVING2, movingRefObj2, tform2, 'OutputView', fixedRefObj, 'SmoothEdges', true); MOVINGREG.RegisteredImage3 = imwarp(MOVING3, movingRefObj3, tform3, 'OutputView', fixedRefObj, 'SmoothEdges', true); MOVINGREG.RegisteredImage4 = imwarp(MOVING4, movingRefObj4, tform4, 'OutputView', fixedRefObj, 'SmoothEdges', true);

% Store spatial referencing object MOVINGREG.SpatialRefObj = fixedRefObj; time= toc; fprintf('The time of execution was %6.2fs\n\n',time);

%Display the image at the end % imshow(MOVINGREG.RegisteredImage1,'Border','tight') figure; % pos1 = [0 -0.1 0.4 1]; % subplot('Position',pos1); imshow(MOVINGREG.RegisteredImage1) % % truesize(MOVINGREG.RegisteredImage1,[100 100]) % pos2 = [0.4 0 0.75 1]; % subplot('Position',pos2); imshow(MOVINGREG.RegisteredImage1) subplot(3,3,4); imshow(MOVINGREG.RegisteredImage1) subplot(3,3,5); imshow(FIXED) subplot(3,3,6); imshow(MOVINGREG.RegisteredImage2) subplot(3,3,8); imshow(MOVINGREG.RegisteredImage3) subplot(3,3,2); imshow(MOVINGREG.RegisteredImage4)

end