def start_menue(self):
"""Starting point of program."""
utils.clear_screen()
options = [
"create NEW classifier", "load EXISTING classifier",
"test parameters", "show high scores", "segment image",
"test data manipulation", "settings"
]
input = utils.menue("START MENUE:", options, True, False)
if input == 1:
self.create_classifier()
elif input == 2:
self.load_classifier()
elif input == 3:
self.test_parameters()
elif input == 4:
self.show_highscores()
elif input == 5:
self.segment_image()
elif input == 6:
self.test_data_menue()
elif input == 7:
self.show_settings()
elif input == 8:
return
self.start_menue()
def show_setting_SURF(self):
utils.clear_screen()
# print current settings
print "SURF SETTINGS"
print "Cross Val. k:\t\t", Settings.SU_CROSS_VALIDATION_K
print "Test Segments:\t\t", Settings.SU_TESTDATA_SEGMENTS
print "SVM Props:\t\t", Settings.SU_SVM_PARAMS
print "BOW Dimension:\t\t", Settings.SU_BOW_DIMENSION
options = ["change cross validation number", "change BOW Dimension"]
input = utils.menue("OPTIONS:", options, False, True)
if input == 1:
Settings.SU_CROSS_VALIDATION_K = utils.value_setting(
"Cross validation runs", "i", "1 means no cross validation.")
self.save_settings()
self.show_setting_SURF()
elif input == 2:
Settings.SU_BOW_DIMENSION = utils.value_setting(
"Change BOW Dimension", "i", "default: 1000")
self.save_settings()
self.show_setting_SURF()
else:
self.show_settings()
def show_setting_histogram(self):
utils.clear_screen()
# print current settings
print "Histogram SETTINGS"
print "Cross Val. k:\t\t", Settings.H_CROSS_VALIDATION_K
print "Test Segments:\t\t", Settings.H_TESTDATA_SEGMENTS
print "Image Segments:\t\t", Settings.H_IMAGE_SEGMENTS
print "Bins per Color\t\t", Settings.H_BINS
print "Range:\t\t\t", Settings.H_COLOR_RANGE
print "Color Space:\t\t", Settings.H_COLOR_SPACE
print "SVM Props:\t\t", Settings.H_SVM_PARAMS
options = [
"change cross validation number", "change image segmentation",
"change number of bins", "change color range", "change color space"
]
input = utils.menue("OPTIONS:", options, False, True)
if input == 1:
Settings.H_CROSS_VALIDATION_K = utils.value_setting(
"Cross validation runs", "i", "1 means no cross validation.")
self.save_settings()
self.show_setting_histogram()
elif input == 2:
Settings.H_IMAGE_SEGMENTS = utils.radio_setting(
"Number of Image Segments", Settings.H_IMAGE_SEGMENTS,
["1", "2", "4", "9", "16", "25"], [1, 2, 4, 9, 16, 25])
self.save_settings()
self.show_setting_histogram()
elif input == 3:
Settings.H_BINS = utils.value_setting(
"Number of histogram bins per color", "i", "default: 64 bins")
self.save_settings()
self.show_setting_histogram()
elif input == 4:
lower = utils.value_setting("Lower range for color values", "i",
"default: 0")
upper = utils.value_setting("Upper range for color values", "i",
"default: 256")
Settings.H_COLOR_RANGE = [lower, upper]
self.save_settings()
self.show_setting_histogram()
elif input == 5:
Settings.H_COLOR_SPACE = utils.radio_setting(
"Color Space", Settings.H_COLOR_SPACE, ["HSV", "RGB", "BGR"],
[ColorSpace.HSV, ColorSpace.RGB, ColorSpace.BGR])
self.save_settings()
self.show_setting_histogram()
else:
self.show_settings()
def show_model_screen(self):
"""
Prints a model selection screen to the console window.
Returns:
ModelId
"""
utils.clear_screen()
options = ["SIFT kp/dp + BoW + SVM", "Histogram & SVM", "SURF kp/dp + BoW + SVM", "ORB kp/dp + BoW + SVM", "DAISY [deprecated]", "CenSurE kp, SIFT dp + BoW + SVM", "Late Fusion - SVM-Vectors", "LBP + SVM", "Neural Nets", "Late Fusion Majority Vote", "Random Keypoint Sampling"]
modelId = utils.menue("Choose model:", options, False)
if modelId == len(options):
return -1
return modelId
def segment_image(self):
""" Menu for image segmentation."""
choice = utils.menue("[Image Segmentation",
["Segment imgage", "Learn custom marker"],
showBackToMain=True)
if choice == 2:
self.learn_marker()
return
elif choice == 3:
return
im = load_image(utils.get_path_via_file_ui(), 1, 1)
im = utils.equalize_image_size(im, 564000)
seg = ObjectSegmentation()
seg.process_image(im)
im = seg.visulize_regions()
cv.imwrite(utils.get_output_path() + "resultsMarker_regions.jpg", im)
try:
plt.imshow(cv.cvtColor(im, cv.COLOR_BGR2RGB), 'gray')
plt.xticks([]), plt.yticks(
[]) # to hide tick values on X and Y axis
plt.show()
utils.save_plt_figure(plt, "image_segments")
except:
print "Could not plot / show image. Saving instead."
utils.save_plt_figure(plt, "image_segments")
colorSeg = ColorSegmentation()
colorSeg.process_image(seg.get_meal_image())
resultImg, images, titles = colorSeg.visulize_regions(seg)
# if matplotlib does not work use cv
for i in range(len(images)):
cv.imwrite(
"C:/Users/felix/Desktop/Segmentation/results/resultsMarker_regionsArea{0}.jpg"
.format(titles[i]), images[i])
cv.imshow("Area {0}".format(titles[i]), images[i])
cv.imshow("Result", resultImg)
cv.imwrite(
"C:/Users/felix/Desktop/Segmentation/results/resultsMarker_resultsImage.jpg",
resultImg)
cv.waitKey(0)
# matplotlib
utils.display_images([resultImg], titles="Result", columns=1, rows=1)
utils.display_images(images, titles=titles)
def show_settings(self):
utils.clear_screen()
# print current settings
print "SETTINGS"
print "DataSet:\t\t", self.show_default_dataset()
print "High Eval Detail:\t", Settings.G_EVALUATION_DETAIL_HIGH
print "Detailed output:\t", Settings.G_DETAILED_CONSOLE_OUTPUT
print "Mail Notifications:\t", MailServer.are_mails_enabled()
print "Auto report mailing:\t", Settings.G_AUTOMATIC_REPORT_MAILING
options = [
"change default data set", "change evaluation detail level",
"change console output level", "SIFT Settings",
"Histogram Settings", "SURF Settings", "Configure Mail Settings"
]
input = utils.menue("OPTIONS:", options, False, True)
if input == 1:
self.change_default_dataset()
self.show_settings()
elif input == 2:
Settings.G_EVALUATION_DETAIL_HIGH = utils.radio_setting(
"Detail Level", Settings.G_EVALUATION_DETAIL_HIGH,
["High", "Low"], [True, False])
self.save_settings()
self.show_settings()
elif input == 3:
Settings.G_DETAILED_CONSOLE_OUTPUT = utils.radio_setting(
"Console Output Detail", Settings.G_DETAILED_CONSOLE_OUTPUT,
["High", "Low"], [True, False])
self.save_settings()
self.show_settings()
elif input == 4:
self.show_setting_SIFT()
self.show_settings()
elif input == 5:
self.show_setting_histogram()
self.show_settings()
elif input == 6:
self.show_setting_SURF()
self.show_settings()
elif input == 7:
self.show_setting_Mail()
self.show_settings()
elif input == 8:
self.start_menue()
def test_data_menue(self):
utils.clear_screen()
options = [
"equalize test data image size", "add prefix to test data",
"crop bounding boxes", "crop to square", "augment test data size",
"cherry pick 'test data augmentation' iteration"
]
input = utils.menue("Test data menue", options, False, True)
if input == 1:
self.normalize_testdata()
elif input == 2:
self.rename_prefix()
elif input == 3:
self.crop_bounding_boxes()
elif input == 4:
self.crop_test_data_to_square()
elif input == 5:
self.augment_test_data()
elif input == 6:
self.augment_test_data(True)
else:
return
self.test_data_menue()
def main_menue(self):
""" Menu that is shown after classifier has been created or loaded."""
# print params of loaded classifier
modelUuid = MODEL_CONTROLLER.model.modelSaver.modelUuid
ml = ModelLoader()
try:
params = ml.get_model_param(modelUuid)
print "\n\nModel type:\t\t", params[0]
print "Changed:\t\t", params[6]
print "Name:\t\t\t", params[1]
print "Model loss:\t\t", params[5]
print "Model id:\t\t", modelUuid
print "Model description:\t", params[2]
print "Trained on:\t\t", params[3]
print "\n"
except:
params = [""]
options = [
"mail reports", "predict random", "generate confusion matrix",
"calculate accuracy results",
"calculate accuracy results and insert score", "show high scores",
"delete model", "predict image folder", "start webcam"
]
if params[0].startswith("NN") or params[0].startswith("CNN"):
options.extend([
"continue training", "plot training history",
"export training history to csv"
])
input = utils.menue("MAIN MENUE:", options)
if input == 1:
self.predict_random_image()
elif input == 2:
self.predict_random_image()
elif input == 3:
self.create_confusion_matrix()
elif input == 4:
self.calculate_results()
raw_input("Press any key to continue.")
elif input == 5:
self.calculate_results(True)
raw_input("Press any key to continue.")
elif input == 6:
self.show_highscores()
elif input == 7:
remove_model(modelUuid)
print "Model removed."
return self.start_menue()
elif input == 8:
self.predict_image_folder()
raw_input("Prediction finished. Press any key to continue.")
elif input == 9:
self.predict_webcam()
raw_input("Prediction finished. Press any key to continue.")
elif input == 10 and (params[0].startswith("NN")
or params[0].startswith("CNN")):
# try to resume training
MODEL_CONTROLLER.run_and_test(False)
elif input == 11 and (params[0].startswith("NN")
or params[0].startswith("CNN")):
MODEL_CONTROLLER.model.plot_training_history()
raw_input("Plotting complete. Press any key to continue.")
elif input == 12 and (params[0].startswith("NN")
or params[0].startswith("CNN")):
MODEL_CONTROLLER.model.export_training_history_to_csv()
raw_input("Exporting complete. Press any key to continue.")
else:
return self.start_menue()
self.main_menue()
def initialize(self):
""" Method to load the classifiers and test them."""
loader = ModelLoader()
input = None
testDataPath = None
while True:
input = utils.menue(
"Majority Vote classifier",
["add classifier", "save", "test + finish", "finish"], False,
True)
if input == 2:
self.modelSaver(self, -1)
continue
if input > 2:
break
# Display loading menu
model = loader.show_loading_screen()
if not model is None:
if testDataPath is None:
testDataPath = model.testData.get_root_path()
self.modelSaver.datasetPath = testDataPath
else:
# check if the test datasets are the same
if testDataPath != model.testData.get_root_path():
print "Could not load classifier {0} because the classifier was trained on different test data.".format(
model.name)
continue
self.classifiers.append(
(1, model)
) # Tuple[0] = model weight (1 is default weight) | Tuple[1] = model
# why did we leave the loop?
if input == 5:
print "Cancel"
return False # cancel -> back to start
else:
# initialize test data for the majority classifier
# check if test data path has changed
if not utils.check_if_dir_exists(testDataPath):
testDataPath = utils.value_question(
"[...]", "Root path to new Dataset Path", "s")
self.testData = TestData(testDataPath, 1, False)
self.testData.segment_test_data({"test": 1})
self.testData.new_segmentation()
self.tester = ModelTester(self)
# test classifier if input == 3
if input == 3:
# Be careful: the results might not reflect the actual accuracy of the classifier.
# if not changed the tester will test on the whole test data set. This might include images that the
# classifiers has been trained on. For a real accuracy test the images have to be separated manually.
results = self.tester.test_classifier(["test"])
self.tester.save_results(results, exportToCSV=False)
print self.tester.format_results_string(results)
testLoss = results["test"][1]
save = utils.radio_question("[?]", "Save/Update classifier?",
None, ["Yes", "No"], [True, False])
if save:
self.modelSaver(self, testLoss)
return self.classifiers != [
] # finish. Return True if classifiers where loaded, False if not.