使用 matlab 为深度学习创建补丁答案

【问题标题】：Creating patches for Deeplearning using matlab使用 matlab 为深度学习创建补丁
【发布时间】：2016-11-13 10:24:48
【问题描述】：

我有一个函数可以为给定的图像创建 32x32 像素的补丁。它返回一个包含所有补丁的单元格。如果图像的格式为 350*350*3，则效果很好，但如果图像的格式为 256*150，则返回带有空图像的单元格。有趣的是，如果我调试它在单元格内创建补丁的代码很好，但是当它在单元格内返回这些补丁时，单元格变为空。我正在尝试使用 setimage 代码保存这些图像。有人可以帮忙吗？

% Demo to divide a color image up into blocks.
function [imageSet] = CreatePatches(imag)
fontSize = 20;
%rgbImage = imread(imag);
rgbImage =imag;
% imshow(rgbImage);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% drawnow;
% Get the dimensions of the image.  numberOfColorBands should be = 3.
[rows columns numberOfColorBands] = size(rgbImage)
%==========================================================================
% divide an image up into blocks is by using mat2cell().
blockSizeR = 32; % Rows in block.
blockSizeC = 32; % Columns in block.
% Figure out the size of each block in rows. 
% Most will be blockSizeR but there may be a remainder amount of less than that.
wholeBlockRows = floor(rows / blockSizeR);
blockVectorR = [blockSizeR * ones(1, wholeBlockRows), rem(rows, blockSizeR)];
% Figure out the size of each block in columns. 
wholeBlockCols = floor(columns / blockSizeC);
blockVectorC = [blockSizeC * ones(1, wholeBlockCols), rem(columns, blockSizeC)];
% Create the cell array, ca.  
% Each cell (except for the remainder cells at the end of the image)
% in the array contains a blockSizeR by blockSizeC by 3 color array.
% This line is where the image is actually divided up into blocks.
if numberOfColorBands > 1
    % It's a color image.
    ca = mat2cell(rgbImage, blockVectorR, blockVectorC, numberOfColorBands);

else
    ca = mat2cell(rgbImage, blockVectorR, blockVectorC);
end
% Now display all the blocks.
plotIndex = 1;
numPlotsR = size(ca, 1);
numPlotsC = size(ca, 2);

for r = 1 : numPlotsR
    for c = 1 : numPlotsC
%       fprintf('plotindex = %d,   c=%d, r=%d\n', plotIndex, c, r);
        % Specify the location for display of the image.
%       subplot(numPlotsR, numPlotsC, plotIndex);
        % Extract the numerical array out of the cell
        % just for tutorial purposes.

        rgbBlock = ca{r,c};
        if mean2(rgbBlock) < 15 % Or whatever value you want
            continue;
        end
%       imshow(rgbBlock); % Could call imshow(ca{r,c}) if you wanted to.
        [rowsB columnsB numberOfColorBandsB] = size(rgbBlock);
        %imwrite(ca{r,c},['image',num2str(plotIndex),'.jpeg']);
        % Make the caption the block number.
%         caption = sprintf('Block #%d of %d\n%d rows by %d columns', ...
%       plotIndex, numPlotsR*numPlotsC, rowsB, columnsB);
%       title(caption);
%       drawnow;
        % Increment the subplot to the next location.
        plotIndex = plotIndex + 1;
         imageSet ={};
         for x =1: plotIndex
             %imshow(rgbBlock);
            imageSet{end+1} = rgbBlock;
         end
    end


end

【问题讨论】：

标签： matlab image-processing deep-learning

【解决方案1】：

只需在检查块的 mean2 值时添加这些行：

if mean2(rgbBlock) < 15|isempty(rgbBlock) == 1

【讨论】：

谢谢，这并不能解释 imageSet 问题，但很有意义并解决了我的问题。