how process sequence of the images in a loop without overwriting

Question

Akhir Shaik 2020-2-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/507367-how-process-sequence-of-the-images-in-a-loop-without-overwriting

评论： Akhir Shaik 2020-3-10

在 MATLAB Online 中打开

here, i am solving around 20000 images at once but the output i'm getting is only the last iamge.

here is the code i got from the Matlab Answers, but also it didn't work can anyone help me.

3 个评论
显示 1更早的评论隐藏 1更早的评论

Akhir Shaik 2020-2-26

here, i need to find the breakup length, need to solve 20k images at once and need to save value in numerical or dat file.

the image processing code is ok. if i solve for 1 images i got the answer but solving for 20k images it is considering only the last image file in the folder.

Stephen23 2020-3-3

在 MATLAB Online 中打开

Original Question: "how process sequence of the images in a loop without overwriting"

here, i am solving around 20000 images at once but the output i'm getting is only the last iamge.

here is the code i got from the Matlab Answers, but also it didn't work can anyone help me.

myFolder = 'E:\IMAGES\100k\flip';
if exist(myFolder, 'dir') ~= 7
  Message = sprintf('Error: The following folder does not exist:\n%s', myFolder);
  uiwait(warndlg(Message));
  return;
end
filePattern = fullfile(myFolder, '*.jpg');
jpegFiles   = dir(filePattern);
for i = 1:length(jpegFiles) 
    fileName = sprintf('%5.5d.jpg', i);
    piccell{i} = imread(fileName);
end
grayImage = imread(fileName);
[rows, columns, numberOfColorChannels] = size(grayImage);
if numberOfColorChannels > 1
	grayImage = rgb2gray(grayImage);
end
highThreshold = 150;
binaryImage = grayImage <= highThreshold;
binaryImage = bwareafilt(binaryImage, 1);
binaryImage = imfill(binaryImage, 'holes');
bottomLine = zeros(rows, 1);
for col = 1 : columns
	thisColumn = binaryImage(:, col);
	lastLine = find(thisColumn, 1, 'last');
	if ~isempty(lastLine)
	bottomLine(col) = lastLine;
	end
end
meanbottomLine = mean(bottomLine(bottomLine>0));
save bkl_.dat meanbottomLine -ascii

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Benjamin Großmann 2020-2-26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/507367-how-process-sequence-of-the-images-in-a-loop-without-overwriting#answer_417439

在 MATLAB Online 中打开

Have a look at your first for loop. Here you are generating a char called fileName which is overwritten in every loop count. furthermore, this loop is unnecessary since you already have your file names in the struct array jpegFiles.

Try to use

fileNameCell = {jpegFiles.name};

to get a cell array of file names. Then write a function which does your image manipulation for one file. Finally, call your function with cellfun and the fileNameCell to apply the image manipulation to every file in the cell.

You can also have a look at the imageSet class.

8 个评论
显示 6更早的评论隐藏 6更早的评论

Benjamin Großmann 2020-2-26

编辑：Benjamin Großmann 2020-2-27

在 MATLAB Online 中打开

First, write the function, save it as findmeanbottomline. This is basically your image manipulation code surrounded by function ... end.

function meanbottomLine =  findmeanbottomline(img_fullfile)
img = imread(img_fullfile);
[rows, columns, numberOfColorChannels] = size(img);
if numberOfColorChannels > 1
    grayImage = rgb2gray(img);
else
    grayImage = img;
end
highThreshold = 150;
binaryImage = grayImage <= highThreshold;
binaryImage = bwareafilt(binaryImage, 1);
binaryImage = imfill(binaryImage, 'holes');
bottomLine = zeros(rows, 1);
for col = 1 : columns
	thisColumn = binaryImage(:, col);
	lastLine = find(thisColumn, 1, 'last');
	if ~isempty(lastLine)
	bottomLine(col) = lastLine;
	end
end
meanbottomLine = mean(bottomLine(bottomLine>0));
end

Then call it with the script

clearvars, close all
clc
myFolder = 'C:\tmp\101_ObjectCategories';
filePattern = fullfile(myFolder, '*/*.jpg'); % my images are in subfolders, so i changed your filepattern
jpegFiles   = dir(filePattern);
% For simplicity of this example, lets use the full/absolute path to the images
fileNameCell = fullfile({jpegFiles.folder}, {jpegFiles.name}); 
% Call the function with cellfun; returns a cell with each entry corresponing to the fileNameCell
meanbottomlineCell = cellfun(@findmeanbottomline, fileNameCell);

Cellfun is archaic, but it's simple and I like it. You can also loop over the fileNameCell instead of using cellfun

for idx = 1:numel(fileNameCell)
    meanbottomline(idx) = findmeanbottomline(fileNameCell{idx});
end

If you have the Parallel Computing Toolbox (and a multi-core processor), you can speed this up by substituting "for" by "parfor" and starting a matlabpool.