import numpy as np
import cv2
import os
import sys
from FeaturesFile import FeaturesFile
from FeaturesLearning import FeaturesLearning
from FileManager import FileManager
from detectAndExtract import detectAndExtract
n_folds = input("Insert the number of folds: ")
detectors_descriptors=[["MSER","SIFT"],["HARRIS","SIFT"],["SIFT","SIFT"],["ORB","ORB"],["FAST","SURF"],["FAST","BRIEF"],["SURF","SURF"]]
for i in range(0,len(detectors_descriptors)):
print "\n######################################"
name = detectors_descriptors[i][0] + "-" + detectors_descriptors[i][1]
print name
ff = FeaturesFile(detectors_descriptors[i][0],detectors_descriptors[i][1])
ff.getFeatures()
X_train, y_train,_ = ff.featuresCategories()
print "#####################"
fl = FeaturesLearning(X_train,y_train,ff)
fl.trainModel(n_folds)
["SIFT", "SIFT"],
["ORB", "ORB"],
["FAST", "SURF"],
["FAST", "BRIEF"],
]
fm = FileManager()
categories = fm.listNoHiddenDir(os.path.dirname(__file__) + os.path.sep + "\\Project_OVA\\imm")
for i in range(0, len(categories)):
currentCategories = categories[i]
print "\n######################################"
print "CURRENT CATEGORY: " + currentCategories
for x in range(0, len(detectors_descriptors)):
print "###########################"
print currentCategories
print "######################################"
print "Features: " + detectors_descriptors[x][0] + " - " + detectors_descriptors[x][1]
ff = FeaturesFile(detectors_descriptors[x][0], detectors_descriptors[x][1], currentCategories)
X_positive, X_negative = ff.getFeatures(currentCategories)
fl = FeaturesLearning(X_positive, X_negative, ff)
fl.trainModel(n_folds)
