How i can extract features from images and save them in one matrix and make groups (label) inside the matrix by matlab?
2 次查看(过去 30 天)
显示 更早的评论
This is works, but i am not sure about my label t_label(1:21,1) is it same rows (From 1 to 21) in the matrix or not,
training_female = 'D:\DBs\jaffe\Train';
filenames = dir(fullfile(training_female, '*.TIFF'));
total_images = numel(filenames);
featureMatrix = [];
for n = 1:total_images
full_name= fullfile(training_female, filenames(n).name);
training_images = imread(full_name);
[featureVector, hogVisualization] = extractHOGFeatures(training_images);
featureMatrix = [featureMatrix; featureVector];
end
xx=imread('YM.SA3.57.TIFF');
[HOG_test, Visualization] = extractHOGFeatures(xx);
t_label = zeros(size(151,1),1);
t_label(1:21,1) = 1;% Angry;
t_label(22:41,1) = 2; %Disgusted;
t_label(42:63,1) = 3; %Fear;
t_label(64:85,1) = 4; %Happy;
t_label(86:106,1) = 5; %neutral;
t_label(107:128,1) = 6; %Sad;
t_label(129:151,1) = 7; %Surprised;
class=knnclassify(HOG_test,featureMatrix,t_label);
2 个评论
Image Analyst
2016-7-23
You might need to give people at least one image so they can try to reproduce your situation.
采纳的回答
Image Analyst
2016-7-23
It looks like rows 1-21 should all have the value 1. Do you know how to look in the workspace to check it? Or just put the cursor in the variable and type control-d.
Or you can just type the variable name in the command window.
0 个评论
更多回答(0 个)
另请参阅
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!