vehicle classification
TRANSCRIPT
VEHICLE CLASSIFICATION
Owais ChishtiZerk ShabanShaleem JohnFaisal UsmanFarman Ali
USING EDGE DETECTION img = imfilter(img, fspecial('laplacian')); Black and white image with edges
img = imfilter(img, fspecial('sobel')); High horizontal edges
img = threshold(img, 200, 0); Apply threshold at 200
img = bwareaopen(img, 30); Remove noise in 30x30 grid
We get a image with shadow highlighted.
USING BAG OF FEATURES What is bag of words/features? Image features as words Occurrence
Processing Steps Feature detection Feature description - SURF Codebook/Dictionary
bagOfFeatures.m Extract image features
Types Find Texture – Similar chunk of images Point of Interest
FREQUENCY IN ARTICLE
GENERATING IMAGE CLASSIFIER SVM – Support Vector Machine Supervised learning Bag of features Plot features to graph Separate features through a line
SOURCE CODE imgSets = imageSet('c:/data-set', 'recursive'); [trainingSets, testSets] = partition(imgSets, 0.3, 'randomize'); bag = bagOfFeatures(trainingSets); categoryClassifier = trainImageCategoryClassifier(trainingSets, bag); confMatrix = evaluate(categoryClassifier, testSets)
Taken from: https://www.mathworks.com/help/vision/ref/trainimagecategoryclassifier.html
REFERENCES https://en.wikipedia.org/wiki/Support_vector_machine https://en.wikipedia.org/wiki/Speeded_up_robust_features https://en.wikipedia.org/wiki/Bag-of-words_model_in_computer_vision
http://www.cs.unc.edu/~lazebnik/spring09/lec18_bag_of_features.pdf