MATLAB Answers

How to check after completing the Image Processing Machine Learning program.

1 view (last 30 days)
Huynh Thanh Sang
Huynh Thanh Sang on 19 Oct 2018
Answered: Bernhard Suhm on 9 Nov 2018
I want to import a picture from the computer to test the program but do not know how.
%%Import Training Data
imgSets = imageSet('folder','recursive');
[imgSets.Count] %show the corresponding count of images
%%Prepare Training Image Sets
minSetCount = min([imgSets.Count]); %determine the smallest a mount of image in a
trainingSets = partition(imgSets, minSetCount, 'randomize');
[trainingSets.Count]
%%Create a Visual Vocabulary from Training Data
bag = bagOfFeatures(trainingSets,'Vocabularysize',100,'Pointselection','Detector');
%%Display Visual Word Occurrence Histograms
% that becomes a new and reduced representation of image.
img = read(imgSets(1), randi(imgSets(1).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,1); imshow(img);
subplot(3,2,2);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
img = read(imgSets(2), randi(imgSets(2).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,3); imshow(img);
subplot(3,2,4);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
img = read(imgSets(3), randi(imgSets(3).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,5); imshow(img);
subplot(3,2,6);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
%%Train a classifier to discriminate between categories
categoryClassifier = trainImageCategoryClassifier(trainingSets, bag); % Train Category Clssifier
Evalute Classifier Performance on Training Set
confMatrix = evaluate(categoryClassifier, trainingSets);

  2 Comments

Huynh Thanh Sang
Huynh Thanh Sang on 19 Oct 2018
%%Import Training Data
imgSets = imageSet('folder','recursive');
[imgSets.Count] %show the corresponding count of images
%%Prepare Training Image Sets
minSetCount = min([imgSets.Count]); %determine the smallest a mount of image in a
trainingSets = partition(imgSets, minSetCount, 'randomize');
[trainingSets.Count]
%%Create a Visual Vocabulary from Training Data
bag = bagOfFeatures(trainingSets,'Vocabularysize',100,'Pointselection','Detector');
%%Display Visual Word Occurrence Histograms
% that becomes a new and reduced representation of image.
img = read(imgSets(1), randi(imgSets(1).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,1); imshow(img);
subplot(3,2,2);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
img = read(imgSets(2), randi(imgSets(2).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,3); imshow(img);
subplot(3,2,4);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
img = read(imgSets(3), randi(imgSets(3).Count));
featureVector = encode(bag, img);
figure(1)
subplot(3,2,5); imshow(img);
subplot(3,2,6);
bar(featureVector); title('Visual Word Occurrences'), xlabel('Visual Word Index'); ylabel('Frequency of occurrences');
%%Train a classifier to discriminate between categories
categoryClassifier = trainImageCategoryClassifier(trainingSets, bag); % Train Category Clssifier
Evalute Classifier Performance on Training Set
confMatrix = evaluate(categoryClassifier, trainingSets);

Sign in to comment.

Answers (1)

Bernhard Suhm
Bernhard Suhm on 9 Nov 2018
Your code isn't complete - it references custom functions you (or your teacher) defined somewhere else such as trainImageCategoryClassifier and evaluate. - Conceptually, you can read an image from your computer using the imread functions, or a test image set by pointing a datastore to the directory with the images. Then you'll want to apply your actually classifier's predict function to that read image of datastore.

  0 Comments

Sign in to comment.

Sign in to answer this question.