def create_mei_files(outdir):
aomr_opts = {
'staff_finder': 0,
'lines_per_staff': 4,
'staff_removal': 0,
'binarization': 0,
'discard_size': 12
}
for dirpath, dirnames, filenames in os.walk(outdir):
if dirpath == outdir:
continue
if ".git" in dirpath.split("/"):
continue
folder_no = os.path.basename(dirpath)
pnum = int(folder_no)
# if not "bad_{0}_corr_page_glyphs.xml".format(folder_no.zfill(4)) in filenames:
# continue
lg.debug("Generating MEI file for {0}".format(pnum))
if "{0}_corr_page_glyphs.xml".format(folder_no.zfill(4)) in filenames:
glyphs = gamera_xml.glyphs_from_xml(os.path.join(dirpath, "{0}_corr_page_glyphs.xml".format(folder_no.zfill(4))))
elif "{0}_uncorr_page_glyphs.xml".format(folder_no.zfill(4)) in filenames:
glyphs = gamera_xml.glyphs_from_xml(os.path.join(dirpath, "{0}_uncorr_page_glyphs.xml".format(folder_no.zfill(4))))
else:
lg.debug("There was no page glyphs file for page {0}".format(pnum))
continue
original_image = os.path.join(dirpath, "{0}_staves_only.tiff".format(folder_no.zfill(4)))
mei_file_write = os.path.join(dirpath, '{0}_uncorr.mei'.format(folder_no.zfill(4)))
aomr_obj = AomrObject(original_image, **aomr_opts)
try:
data = aomr_obj.run(glyphs)
except OverflowError, e:
lg.debug("Could not do detection on {0} because {1}".format(pnum, e))
continue
if not data:
# no data was returned.
lg.debug("No data was found for {0}".format(pnum))
continue
mei_file = AomrMeiOutput.AomrMeiOutput(data, "{0}_original_image.tiff".format(pnum), page_number = pnum)
meitoxml.meitoxml(mei_file.md, mei_file_write)
def main(options):
init_gamera()
all_glyphs = []
for dirpath, dirnames, filenames in os.walk(options['gam']):
for f in filenames:
if f != "page_glyphs.xml":
continue
# lg.debug("Processing: {0}".format(f))
glyphs = gamera_xml.glyphs_from_xml(os.path.join(dirpath, f))
def __glyphchecker(g):
if "_group" in g.get_main_id().split("."):
return False
elif "_split" in g.get_main_id().split("."):
return False
else:
return True
all_glyphs.extend([g.get_main_id() for g in glyphs if __glyphchecker(g)])
all_glyphs.sort()
all_glyph_set = set(all_glyphs)
all_glyph_list = list(all_glyph_set)
all_glyph_list.sort()
for n in all_glyph_list:
print n
def pitch_find(original_file, page_file, outdir, pitch_find_algorithm, exceptions):
aomr_opts = {
'lines_per_staff': 4,
'staff_finder': 0, # 0: Miyao
'staff_removal': 0,
'binarization': 0,
'discard_size': 12,
'exceptions': exceptions
}
#FILES TO PROCESS
glyphs = gamera_xml.glyphs_from_xml(page_file)
file_name = (original_file.split('/')[-2] + '_' + original_file.split('/')[-1])
# CREATING AOMR OBJECT, FINDING STAVES, AND RETRIEVING STAFF COORDINATES
aomr_obj = AomrObject(original_file, **aomr_opts)
st_position = aomr_obj.find_staves() # staves position
staff_coords = aomr_obj.staff_coords()
las = proc_st_pos(st_position)
# staff_non_parallel = aomr_obj.staff_no_non_parallel(glyphs)
# print staff_non_parallel
if pitch_find_algorithm == 'Miyao':
sorted_glyphs = aomr_obj.miyao_pitch_find(glyphs, aomr_opts['discard_size'])
elif pitch_find_algorithm == 'AvLines':
sorted_glyphs = aomr_obj.pitch_find(glyphs, st_position, aomr_opts['discard_size'])
return sorted_glyphs, st_position, aomr_obj, las
def run_my_task(self, inputs, settings, outputs):
classifier_path = inputs['GameraXML - Training Data'][0][
'resource_path']
with self.tempdir() as tdir:
tempPath = os.path.join(tdir, classifier_path + '.xml')
copyfile(classifier_path, tempPath)
cknn = gamera.knn.kNNNonInteractive(tempPath)
if 'GameraXML - Feature Selection' in inputs:
cknn.load_settings(
inputs['GameraXML - Feature Selection'][0]['resource_path'])
func = gamera.classify.BoundingBoxGroupingFunction(
settings['Bounding box size'])
# Load the connected components
ccs = glyphs_from_xml(
inputs['GameraXML - Connected Components'][0]['resource_path'])
# Do grouping
cs_image = cknn.group_and_update_list_automatic(ccs,
grouping_function=func,
max_parts_per_group=4,
max_graph_size=16)
# Generate the Gamera features
cknn.generate_features_on_glyphs(cs_image)
# Write the glyphs to GameraXML
output_xml = gamera.gamera_xml.WriteXMLFile(glyphs=cs_image,
with_features=True)
for i in range(len(outputs['GameraXML - Classified Glyphs'])):
output_xml.write_filename(
outputs['GameraXML - Classified Glyphs'][i]['resource_path'])
def process_glyphs_directory(glyphs_directory, output_dir):
aomr_opts = {
'staff_finder': 0,
'lines_per_staff': 4,
'staff_removal': 0,
'binarization': 0,
'discard_size': 12
}
print "Processing glyphs directory"
for dirpath, dirnames, filenames in os.walk(glyphs_directory):
for f in filenames:
if f == 'page_glyphs.xml':
folder_no = os.path.basename(dirpath)
pnum = int(folder_no)
input_filename = os.path.join(dirpath, f)
lg.debug("Input filename is {0}".format(input_filename))
output_filename = os.path.join(output_dir, folder_no.zfill(4), f)
lg.debug("Output filename is {0}".format(output_filename))
shutil.copy(input_filename, output_filename)
original_image = os.path.join(output_dir, folder_no.zfill(4), 'original_image.tiff')
mei_file_write = os.path.join(output_dir, folder_no.zfill(4), 'liber-usualis-{0}.mei'.format(folder_no.zfill(4)))
glyphs = gamera_xml.glyphs_from_xml(output_filename)
aomr_obj = AomrObject(original_image, **aomr_opts)
data = aomr_obj.run(glyphs)
mei_file = AomrMeiOutput.AomrMeiOutput(data, original_image.split('/')[-2], page_number = pnum)
meitoxml.meitoxml(mei_file.md, mei_file_write)
def run_my_task(self, inputs, settings, outputs):
classifier_path = inputs['GameraXML - Training Data'][0]['resource_path']
with self.tempdir() as tdir:
tempPath = os.path.join(tdir, classifier_path + '.xml')
copyfile(classifier_path, tempPath)
cknn = gamera.knn.kNNNonInteractive(tempPath)
if 'GameraXML - Feature Selection' in inputs:
cknn.load_settings(inputs['GameraXML - Feature Selection'][0]['resource_path'])
func = gamera.classify.BoundingBoxGroupingFunction(
settings['Bounding box size'])
# Load the connected components
ccs = glyphs_from_xml(
inputs['GameraXML - Connected Components'][0]['resource_path'])
# Do grouping
cs_image = cknn.group_and_update_list_automatic(ccs,
grouping_function=func,
max_parts_per_group=4,
max_graph_size=16)
# Generate the Gamera features
cknn.generate_features_on_glyphs(cs_image)
# Write the glyphs to GameraXML
output_xml = gamera.gamera_xml.WriteXMLFile(glyphs=cs_image,
with_features=True)
for i in range(len(outputs['GameraXML - Classified Glyphs'])):
output_xml.write_filename(
outputs['GameraXML - Classified Glyphs'][i]['resource_path'])
def main(original_file, page_file, outdir):
aomr_opts = {
'lines_per_staff': 4,
'staff_finder': 0, # 0: Miyao
'staff_removal': 0,
'binarization': 0,
'discard_size': 12
}
#FILES TO PROCESS
glyphs = gamera_xml.glyphs_from_xml(page_file)
lg.debug("Original file: {0}, page file: {1}".format(original_file, page_file))
page_number = int(original_file.split('/')[-2])
fname = os.path.splitext(original_file.split('/')[-1])[0] + ".mei"
file_name = "{0}_{1}".format(page_number, fname)
# CREATING AOMR OBJECT, FINDING STAVES, AND RETRIEVING STAFF COORDINATES
aomr_obj = AomrObject(original_file, **aomr_opts)
st_position = aomr_obj.find_staves() # staves position
staff_coords = aomr_obj.staff_coords()
sorted_glyphs = aomr_obj.miyao_pitch_find(glyphs)
# PITCH FINDING
# pitch_find = aomr_obj.pitch_find(glyphs, st_position, aomr_opts.get('discard_size'))
# print len(pitch_find)
# sorted_glyphs = sorted(proc_glyphs, key=itemgetter(1, 2))
# STRUCTURING THE DATA IN JSON
data = {}
for s, stave in enumerate(staff_coords):
contents = []
for glyph, staff, offset, strt_pos, note, octave, clef_pos, clef in sorted_glyphs:
glyph_id = glyph.get_main_id()
# lg.debug("Glyph ID: {0}".format(glyph_id))
glyph_type = glyph_id.split(".")[0]
glyph_form = glyph_id.split(".")[1:]
# lg.debug("sg[1]:{0} s:{1} sg{2}".format(sg[1], s+1, sg))
# structure: g, stave, g.offset_x, note, strt_pos
if staff == s+1:
j_glyph = { 'type': glyph_type,
'form': glyph_form,
'coord': [glyph.offset_x, glyph.offset_y, glyph.offset_x + glyph.ncols, glyph.offset_y + glyph.nrows],
'strt_pitch': note,
'octv': octave,
'strt_pos': strt_pos,
'clef_pos': clef_pos,
'clef': clef}
contents.append(j_glyph)
data[s] = {'coord':stave, 'content':contents}
#print data
# CREATING THE MEI FILE
mei_file = AomrMeiOutput.AomrMeiOutput(data, file_name, page_number)
meitoxml.meitoxml(mei_file.md, os.path.join(outdir, file_name))
def test_noninteractive_classifier():
# We assume the XML reading/writing itself is fine (given
# test_xml), but we should test the wrappers in classify anyway
image = load_image("data/testline.png")
ccs = image.cc_analysis()
database = gamera_xml.glyphs_from_xml("data/testline.xml")
classifier = knn.kNNNonInteractive(database,features=featureset,normalize=False)
assert not classifier.is_interactive()
assert len(classifier.get_glyphs()) == 66
_test_classification(classifier, ccs)
classifier.serialize("tmp/serialized.knn")
classifier.clear_glyphs()
assert len(classifier.get_glyphs()) == 0
classifier.unserialize("tmp/serialized.knn")
def home(request):
init_gamera()
encoded_glyphs = []
gamera_glyphs = gamera_xml.glyphs_from_xml(
'/Volumes/Shared/LU-OMR/Liber_Usualis_NO_ST/Processed_Pages/1234/classifier_glyphs_1234_654.xml'
)
for gamera_glyph in gamera_glyphs:
glyph = gamera_glyph.to_rgb().to_pil()
buf = StringIO.StringIO()
glyph.save(buf, format='PNG')
png = buf.getvalue()
encoded_png = base64.b64encode(png)
encoded_glyphs.append(encoded_png)
for glyph in encoded_glyphs:
print glyph
print ""
return render(request, 'index.html', {'encoded_glyphs': encoded_glyphs})
def home(request):
init_gamera()
encoded_glyphs = []
gamera_glyphs = gamera_xml.glyphs_from_xml('/Volumes/Shared/LU-OMR/Liber_Usualis_NO_ST/Processed_Pages/1234/classifier_glyphs_1234_654.xml')
for gamera_glyph in gamera_glyphs:
glyph = gamera_glyph.to_rgb().to_pil()
buf = StringIO.StringIO()
glyph.save(buf, format='PNG')
png = buf.getvalue()
encoded_png = base64.b64encode(png)
encoded_glyphs.append(encoded_png)
for glyph in encoded_glyphs:
print glyph
print ""
return render(request, 'index.html', {
'encoded_glyphs': encoded_glyphs
})
def test_noninteractive_classifier():
# We assume the XML reading/writing itself is fine (given
# test_xml), but we should test the wrappers in classify anyway
image = load_image("data/testline.png")
ccs = image.cc_analysis()
database = gamera_xml.glyphs_from_xml("data/testline.xml")
classifier = knn.kNNNonInteractive(database,
features=featureset,
normalize=False)
assert not classifier.is_interactive()
assert len(classifier.get_glyphs()) == 66
_test_classification(classifier, ccs)
classifier.serialize("tmp/serialized.knn")
classifier.clear_glyphs()
assert len(classifier.get_glyphs()) == 0
classifier.unserialize("tmp/serialized.knn")
def run_my_task(self, inputs, settings, outputs):
glyphs = gamera_xml.glyphs_from_xml(
inputs['GameraXML - Connected Components'][0]['resource_path'])
print settings
kwargs = {
'smoothing': settings['Smoothing'],
'extrema_threshold': 0,
'min_glyph_size': settings['Minimum Glyph Size'],
'max_recursive_cuts': settings['Maximum Recursive Cuts'],
'rotation': settings['Angle'],
# will it cut?
'min_slice_spread_rel':
settings['Minimum Slice Spread'], # minimum spread for a cut
'low_projection_threshold': settings[
'Low Valley Threshold'], # FORCE a cut if valley under a certain value
'min_projection_segments': settings[
'Minimum Segment Length'], # ++ less likely to cut, -- more slice points
'prefer_multi_cuts': False,
'prefer_x': settings['Prioritize X-slices'],
# Debug Options
'print_projection_array': False,
'plot_projection_array': False, # script only
'save_cuts': False,
}
ps = ProjectionSplitter(**kwargs)
output_glyphs = []
for g in glyphs:
output_glyphs += ps.run(g)
outfile_path = outputs['GameraXML - Connected Components'][0][
'resource_path']
output_xml = gamera_xml.WriteXMLFile(glyphs=output_glyphs,
with_features=True)
output_xml.write_filename(outfile_path)
return True
def main(options):
init_gamera()
glyph_corrections = dict(all_glyphs.g)
errno = 0.
numglyphs = 0.
numpages = 0
for dirpath, dirnames, filenames in os.walk(options['gam']):
numpages += 1
for f in filenames:
if f == "page_glyphs_corr.xml":
os.unlink(os.path.join(dirpath, f))
if f != "page_glyphs.xml":
continue
# lg.debug("Processing: {0}".format(f))
glyphs = gamera_xml.glyphs_from_xml(os.path.join(dirpath, f))
for i,g in enumerate(glyphs):
numglyphs += 1
if g.get_main_id() in glyph_corrections.keys():
errno += 1
# this glyph needs correcting.
action = glyph_corrections[g.get_main_id()]
if action == "warn":
lg.warn("=====> Problem detected with {0} on page {1} <=====".format(g.get_main_id(), os.path.basename(dirpath)))
elif action == "":
# lg.warn("Deleting {0} from page {1}".format(g.get_main_id(), dirpath))
glyphs.pop(i)
else:
# lg.warn("Replacing {0} with {1} on page {2}".format(g.get_main_id(), action, dirpath))
g.classify_manual(action)
os.rename(os.path.join(dirpath, f), os.path.join(dirpath, "page_glyphs_uncorr.xml"))
gamera_xml.WriteXMLFile(glyphs=glyphs, with_features=True).write_filename(os.path.join(dirpath, "page_glyphs.xml"))
# save out glyphs.
# move on.
print "Totals: {0} errors out of {1} glyphs; {2} total error rate".format(errno, numglyphs, (errno / numglyphs) * 100)
print "Number of pages: {0}".format(numpages)
print "Average glyphs per page: {0}".format(numglyphs / numpages)
print "Average errors per page: {0}".format(errno / numpages)
def run_my_task(self, inputs, settings, outputs):
# Initialize a gamera classifier
classifier_path = inputs['GameraXML - Connected Components'][0][
'resource_path']
if 'GameraXML - Feature Selection' in inputs:
features = inputs['GameraXML - Feature Selection'][0][
'resource_path']
else:
features = None
# Handle importing the optional training classifier
if 'GameraXML - Training Data' in inputs:
training_database = glyphs_from_xml(
inputs['GameraXML - Training Data'][0]['resource_path'])
else:
training_database = []
# Set the initial state
if '@state' not in settings:
settings['@state'] = ClassifierStateEnum.IMPORT_XML
settings['glyphs'] = []
if settings['@state'] == ClassifierStateEnum.IMPORT_XML:
run_import_stage(settings, classifier_path)
settings['@state'] = ClassifierStateEnum.CORRECTION
settings['glyphs_json'] = serialize_glyphs_to_json(settings)
return self.WAITING_FOR_INPUT()
elif settings['@state'] == ClassifierStateEnum.CORRECTION:
# Update any changed glyphs
update_changed_glyphs(settings)
run_correction_stage(settings, training_database, features)
settings['glyphs_json'] = serialize_glyphs_to_json(settings)
return self.WAITING_FOR_INPUT()
else:
# Update changed glyphs
update_changed_glyphs(settings)
# Do one final classification before quitting
cknn = run_correction_stage(settings, training_database, features)
# No more corrections are required. We can now output the data
run_output_stage(cknn, settings['glyphs'], outputs)
# Remove the JSON string from the database
purge_serialized_json(settings)
def run_my_task(self, inputs, settings, outputs):
# Inputs
infile_path = inputs['JSOMR of staves and page properties'][0][
'resource_path']
with open(infile_path, 'r') as infile:
jsomr_string = infile.read()
jsomr = json.loads(jsomr_string)
glyphs = gamera_xml.glyphs_from_xml(
inputs['GameraXML - Classified Connected Components'][0]
['resource_path'])
kwargs = {
'discard_size': settings['Discard Size'],
}
pf = PitchFinder(**kwargs)
page = jsomr['page']
staves = jsomr['staves']
pitches = pf.get_pitches(glyphs, staves)
# Outputs
jsomr = {
'page': page,
'staves': staves,
'glyphs': pitches,
}
outfile_path = outputs['JSOMR of glyphs, staves, and page properties'][
0]['resource_path']
with open(outfile_path, 'w') as outfile:
outfile.write(json.dumps(jsomr))
return True
def fix_spelling_mistakes(page_glyphs_file, glyph_corrections):
errno = 0.
numglyphs = 0.
numpages = 0
glyphs = gamera_xml.glyphs_from_xml(page_glyphs_file)
k = glyph_corrections.keys()
for i,g in enumerate(glyphs):
numglyphs += 1
if g.get_main_id() in k:
errno += 1
# this glyph needs correcting.
action = glyph_corrections[g.get_main_id()]
if action == "warn":
lg.warn("=====> Problem detected with {0} on page {1} <=====".format(g.get_main_id(), os.path.basename(page_glyphs_file)))
elif action == "":
# lg.warn("Deleting {0} from page {1}".format(g.get_main_id(), dirpath))
glyphs.pop(i)
else:
# lg.warn("Replacing {0} with {1} on page {2}".format(g.get_main_id(), action, dirpath))
g.classify_manual(action)
return (glyphs, numglyphs, errno)
def test():
glyphs = gamera_xml.glyphs_from_xml(
r"C:\Documents and Settings\Karl MacMillan\Desktop\test\prod.xml")
glyphs = strip_small_categories(glyphs)
from gamera.plugins import features
k = knn.kNN()
print(k.features)
features.generate_features_list(glyphs, k.feature_functions)
print("Getting gstats")
graph_stats = get_graph_stats(glyphs, k)
gstats = knn.get_glyphs_stats(glyphs)
max_dist = max_distance(glyphs, k)
print(max_dist)
file = open("results.txt", "w")
global_max = [[], []]
local_max = [[], []]
all = [[], []]
graph = [[], []]
gr_ccorrect = 0
gr_icorrect = 0
for x in glyphs:
local_max_dist = local_max_distance(glyphs, x, k)
ans = k.classify_with_images(glyphs, x, 1)
file.write(ans[0][1] + ",") # + str(ans[0][0]) + ",")
correct = 0
if x.get_main_id() == ans[0][1]:
file.write("1,")
correct = 1
else:
file.write("0,")
g = 1.0 - (ans[0][0] / max_dist)
global_max[correct].append(g)
file.write(str(g) + ",")
l = 1.0 - (ans[0][0] / local_max_dist)
local_max[correct].append(l)
file.write(str(l) + ",")
a = stats.samplestdev([ans[0][0], gstats[ans[0][1]][1]])
all[correct].append(a)
file.write(str(a) + ",")
gr = stats.samplestdev([ans[0][0], graph_stats[ans[0][1]]])
if (gr <= 1 and correct):
gr_ccorrect += 1
if (gr > 1 and not correct):
gr_icorrect += 1
graph[correct].append(gr)
file.write(str(gr))
file.write("\n")
print("num correct: %d num incorrect: %d" %
(len(global_max[1]), len(global_max[0])))
print("confidence %f %f %f" %
(((gr_ccorrect + gr_icorrect) / float(len(glyphs))),
gr_ccorrect / float(len(glyphs) - len(global_max[0])),
gr_icorrect / float(len(glyphs) - len(global_max[1]))))
cgm = -1
igm = -1
cgs = -1
igs = -1
if (len(global_max[0])):
igm = stats.mean(global_max[0])
igs = stats.samplestdev(global_max[0])
if (len(global_max[1])):
cgm = stats.mean(global_max[1])
cgs = stats.samplestdev(global_max[1])
clm = -1
ilm = -1
cls = -1
ils = -1
if (len(local_max[0])):
ilm = stats.mean(local_max[0])
ils = stats.samplestdev(local_max[0])
if (len(local_max[1])):
clm = stats.mean(local_max[1])
cls = stats.samplestdev(local_max[1])
cam = -1
iam = -1
cas = -1
ias = -1
if (len(all[0])):
iam = stats.mean(all[0])
ias = stats.samplestdev(all[0])
if (len(all[1])):
cam = stats.mean(all[1])
cas = stats.samplestdev(all[1])
cgraphm = -1
igraphm = -1
cgraphs = -1
igraphs = -1
if (len(graph[0])):
igraphm = stats.mean(graph[0])
igraphs = stats.samplestdev(graph[0])
if (len(graph[1])):
cgraphm = stats.mean(graph[1])
cgraphs = stats.samplestdev(graph[1])
print("global correct avg: %f stdev: %f incorrect avg: %f stddev: %f" %
(cgm, cgs, igm, igs))
print("local correct avg: %f stdev: %f incorrect avg: %f stddev: %f" %
(clm, cls, ilm, ils))
print("all correct avg: %f stdev: %f incorrect avg: %f stddev: %f" %
(cam, cas, iam, ias))
print("graph correct avg: %f stdev: %f incorrect avg: %f stddev: %f" %
(cgraphm, cgraphs, igraphm, igraphs))
def otsu_threshold(p):
l = len(p)
mu_T = 0.0
for i in range(l):
mu_T += i * p[i]
sigma_T = 0.0
for i in range(l):
sigma_T += (i - mu_T) * (i - mu_T) * p[i]
k_low = 0
while (p[k_low] == 0) and (k_low < (l - 1)):
k_low += 1
k_high = l - 1
while (p[k_high] == 0) and (k_high > 0):
k_low += 1
k_high -= 1
criterion = 0.0
thresh = 127
omega_k = 0.0
mu_k = 0.0
k = k_low
while k <= k_high:
omega_k += p[k]
mu_k += k * p[k]
expr_1 = (mu_T * omega_k - mu_k)
sigma_b_k = expr_1 * expr_1 / (omega_k * (1 - omega_k))
if (criterion < sigma_b_k / sigma_T):
criterion = sigma_b_k / sigma_T
thresh = k
k += 1
return thresh
graph_l = graph[0][:]
graph_l.extend(graph[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print("threshold: " + str(threshold))
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[0])) * 100)
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[1])) * 100)
graph_l = all[0][:]
graph_l.extend(all[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print("threshold: " + str(threshold))
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[0])) * 100)
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[1])) * 100)
graph_l = local_max[0][:]
graph_l.extend(local_max[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print("threshold: " + str(threshold))
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[0])) * 100)
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print(num_wrong, num_wrong / float(len(graph[1])) * 100)
def test_glyphs_with_features_from_xml():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml",
["area", "aspect_ratio"])
assert len(glyphs) == 66
assert len(glyphs[0].features) == 2
from gamera.toolkits.aruspix.ax_file import *
axfile = ""
group = 0
# start options dialog
dialog = Args(
[FileOpen("Aruspix file", axfile, "*.axz"), Choice("group", ["Group", "No group"])], name="Select the file"
)
params = dialog.show()
if params is not None and params[0]:
# map parameters
i = 0
axfile = params[i]
i += 1
group = params[i]
i += 1
f = AxFile(axfile, "")
gl = []
if group == 1:
gl = gamera_xml.glyphs_from_xml(f.tmpdirname + "gamera_page_no_group.xml")
else:
gl = gamera_xml.glyphs_from_xml(f.tmpdirname + "gamera_page_group.xml")
image = load_image(f.tmpdirname + "img2.tif")
classifier = knn.kNNInteractive()
classifier.display(gl, image)
def test_glyphs_to_xml_with_features():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml",
feature_functions=features)
gamera_xml.glyphs_to_xml("tmp/testline_test2.xml", glyphs, True)
assert equal_files("tmp/testline_test2.xml", "data/testline_test2.xml")
def test_glyphs_to_xml_gz():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml")
gamera_xml.glyphs_to_xml("tmp/testline_test1.xml.gz", glyphs, False)
assert equal_files("tmp/testline_test1.xml.gz", "data/testline_test1.xml.gz", gz=True)
l, h = image.ncols, image.nrows
p = int(0.5 + l / 2.0), int(0.5 + h / 2.0)
return p
if __name__ == "__main__":
inImage, inXML = None, None
(in0) = sys.argv[1]
if '.png' in in0:
inImage = in0
image = load_image(inImage)
elif '.xml' in in0:
inXML = in0
glyphs = gamera_xml.glyphs_from_xml(inXML)
# remove files already there so they dont get stacked up
filesPNG = glob.glob('./output/*.png')
filesXML = glob.glob('./output/*.xml')
for f in filesPNG + filesXML:
os.remove(f)
kwargs = {
'smoothing': 1,
'min_glyph_size': 20,
'max_recursive_cuts': 50,
'rotation': 45,
# will it cut?
'min_slice_spread_rel': 0.2, # minimum spread for a cut
def test_glyphs_from_xml_gz():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml.gz")
assert len(glyphs) == 66
def _test_malformed():
glyphs = gamera_xml.glyphs_from_xml("data/malformed.xml")
def _test_missing_attributes():
glyphs = gamera_xml.glyphs_from_xml("data/missing_attributes.xml")
def _test_malformed():
glyphs = gamera_xml.glyphs_from_xml("data/malformed.xml")
def test_write_xml():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml")
writer = gamera_xml.WriteXML(glyphs)
result_string = writer.string()
writer.write_filename("tmp/testline_test3.xml")
assert equal_files("tmp/testline_test3.xml", "data/testline_test3.xml")
def __init__(self, gamera_file_path):
self.gamera_images = glyphs_from_xml(gamera_file_path)
# Construct the glyph objects
self.glyphs = [gamera_image_to_glyph(image) for image in
self.gamera_images]
def test_glyphs_to_xml_gz():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml")
gamera_xml.glyphs_to_xml("tmp/testline_test1.xml.gz", glyphs, False)
assert equal_files("tmp/testline_test1.xml.gz", "data/testline_test1.xml.gz", gz=True)
def test():
glyphs = gamera_xml.glyphs_from_xml(r"C:\Documents and Settings\Karl MacMillan\Desktop\test\prod.xml")
glyphs = strip_small_categories(glyphs)
from gamera.plugins import features
k = knn.kNN()
print k.features
features.generate_features_list(glyphs, k.feature_functions)
print "Getting gstats"
graph_stats = get_graph_stats(glyphs, k)
gstats = knn.get_glyphs_stats(glyphs)
max_dist = max_distance(glyphs, k)
print max_dist
file = open("results.txt", "w")
global_max = [[],[]]
local_max = [[],[]]
all = [[],[]]
graph = [[],[]]
gr_ccorrect = 0
gr_icorrect = 0
for x in glyphs:
local_max_dist = local_max_distance(glyphs, x, k)
ans = k.classify_with_images(glyphs, x, 1)
file.write(ans[0][1] + ",")# + str(ans[0][0]) + ",")
correct = 0
if x.get_main_id() == ans[0][1]:
file.write("1,")
correct = 1
else:
file.write("0,")
g = 1.0 - (ans[0][0] / max_dist)
global_max[correct].append(g)
file.write(str(g) + ",")
l = 1.0 - (ans[0][0] / local_max_dist)
local_max[correct].append(l)
file.write(str(l) + ",")
a = stats.samplestdev([ans[0][0],gstats[ans[0][1]][1]])
all[correct].append(a)
file.write(str(a) + ",")
gr = stats.samplestdev([ans[0][0],graph_stats[ans[0][1]]])
if (gr <= 1 and correct):
gr_ccorrect += 1
if (gr > 1 and not correct):
gr_icorrect += 1
graph[correct].append(gr)
file.write(str(gr))
file.write("\n")
print "num correct: %d num incorrect: %d" % (len(global_max[1]), len(global_max[0]))
print "confidence %f %f %f" % (((gr_ccorrect + gr_icorrect) / float(len(glyphs))),
gr_ccorrect / float(len(glyphs) - len(global_max[0])),
gr_icorrect / float(len(glyphs) - len(global_max[1])))
cgm = -1
igm = -1
cgs = -1
igs = -1
if (len(global_max[0])):
igm = stats.mean(global_max[0])
igs = stats.samplestdev(global_max[0])
if (len(global_max[1])):
cgm = stats.mean(global_max[1])
cgs = stats.samplestdev(global_max[1])
clm = -1
ilm = -1
cls = -1
ils = -1
if (len(local_max[0])):
ilm = stats.mean(local_max[0])
ils = stats.samplestdev(local_max[0])
if (len(local_max[1])):
clm = stats.mean(local_max[1])
cls = stats.samplestdev(local_max[1])
cam = -1
iam = -1
cas = -1
ias = -1
if (len(all[0])):
iam = stats.mean(all[0])
ias = stats.samplestdev(all[0])
if (len(all[1])):
cam = stats.mean(all[1])
cas = stats.samplestdev(all[1])
cgraphm = -1
igraphm = -1
cgraphs = -1
igraphs = -1
if (len(graph[0])):
igraphm = stats.mean(graph[0])
igraphs = stats.samplestdev(graph[0])
if (len(graph[1])):
cgraphm = stats.mean(graph[1])
cgraphs = stats.samplestdev(graph[1])
print "global correct avg: %f stdev: %f incorrect avg: %f stddev: %f" % (cgm, cgs, igm, igs)
print "local correct avg: %f stdev: %f incorrect avg: %f stddev: %f" % (clm, cls, ilm, ils)
print "all correct avg: %f stdev: %f incorrect avg: %f stddev: %f" % (cam, cas, iam, ias)
print "graph correct avg: %f stdev: %f incorrect avg: %f stddev: %f" % (cgraphm, cgraphs, igraphm, igraphs)
def otsu_threshold(p):
l = len(p)
mu_T = 0.0
for i in range(l):
mu_T += i * p[i]
sigma_T = 0.0
for i in range(l):
sigma_T += (i-mu_T)*(i-mu_T)*p[i]
k_low = 0
while (p[k_low] == 0) and (k_low < (l - 1)):
k_low += 1
k_high = l - 1
while (p[k_high] == 0) and (k_high > 0):
k_low += 1
k_high -= 1
criterion = 0.0
thresh = 127
omega_k = 0.0
mu_k = 0.0
k = k_low
while k <= k_high:
omega_k += p[k]
mu_k += k*p[k]
expr_1 = (mu_T*omega_k - mu_k)
sigma_b_k = expr_1 * expr_1 / (omega_k*(1-omega_k))
if (criterion < sigma_b_k/sigma_T):
criterion = sigma_b_k/sigma_T
thresh = k;
k += 1
return thresh
graph_l = graph[0][:]
graph_l.extend(graph[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print "threshold: " + str(threshold)
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[0])) * 100
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[1])) * 100
graph_l = all[0][:]
graph_l.extend(all[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print "threshold: " + str(threshold)
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[0])) * 100
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[1])) * 100
graph_l = local_max[0][:]
graph_l.extend(local_max[1])
graph_l.sort()
threshold = stats.mean(graph_l)
print "threshold: " + str(threshold)
num_wrong = 0
for x in graph[0]:
if x < threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[0])) * 100
num_wrong = 0
for x in graph[1]:
if x >= threshold:
num_wrong += 1
print num_wrong, num_wrong / float(len(graph[1])) * 100
def test_write_xml():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml")
writer = gamera_xml.WriteXML(glyphs)
result_string = writer.string()
writer.write_filename("tmp/testline_test3.xml")
assert equal_files("tmp/testline_test3.xml", "data/testline_test3.xml")
def _test_missing_attributes():
glyphs = gamera_xml.glyphs_from_xml("data/missing_attributes.xml")
def test_glyphs_from_xml_gz():
glyphs = gamera_xml.glyphs_from_xml("data/testline.xml.gz")
assert len(glyphs) == 66
