# image_data = phyloParser.removeRepeatedLines(image_data)
image_data = phyloParser.groupLines(image_data, debug=False)
image_data = phyloParser.matchLineGroups(image_data, debug=False)
# # image_data = phyloParser.detectCorners(image_data)
# # image_data = phyloParser.refineLinesByCorners(image_data)
# # image_data = phyloPaser.getSpecies_v2(image_data, debug = True)
# # image_data = phyloPaser.getSpecies(image_data, debug = True)
# image_data = phyloParser.matchLines(image_data, debug = True, useNew = False)
# image_data = phyloParser.getSpecies_v2(image_data, debug = True)
# # print image_data.orphanBox2Text
image_data = phyloParser.getSpecies_v3(image_data, debug=False)
# testString = phyloParser.makeTree(image_data, debug = True, tracing = False)
# treeString = phyloParser.constructTreeByTracing(image_data, debug = False)
treeString = phyloParser.constructTree_eval(image_data,
debug=False,
tracing=False)
print treeString.treeStructure
dbdata = treeString.treeHead.getParentChildrenStyle()
print dbdata
# print dbdata['cluster']
with open('dbdata_2.csv', 'ab') as f:
csvwriter = csv.writer(f, delimiter='\t')
for i in range(dbdata['relation_index'] + 1):
if i >= dbdata['count_labels']:
string = ''
for node in dbdata[i]:
string += '#' + str(node)
csvwriter.writerow([fileName, str(i), string])
image_data = phyloParser.detectLines(image_data, debug = False)
image_data = phyloParser.getCorners(image_data, debug = False)
image_data = phyloParser.makeLinesFromCorner(image_data, debug = False)
image_data = phyloParser.includeLinesFromCorners(image_data)
image_data = phyloParser.postProcessLines(image_data)
image_data = phyloParser.groupLines(image_data)
image_data = phyloParser.matchLineGroups(image_data, debug = False)
image_data = phyloParser.getSpecies_v3(image_data, debug = False)
image_data = phyloParser.constructTree_eval(image_data, fixTree = False, tracing = False, orphanHint = False, debug = False)
####### evaluation
truth_string = string2TreeString(ground_truth[fileNameList[index]], rename = True)
print image_data.treeStructure
t1 = PhyloTree(image_data.treeStructure + ";")
t2 = PhyloTree(truth_string + ";")
PhyloTree.rename_node(t1, rename_all=True)
PhyloTree.rename_node(t2, rename_all=True)
distance = PhyloTree.zhang_shasha_distance(t1, t2)
num_node = PhyloTree.getNodeNum(t2)
num_leaf = t2.getLeafCount()
score = distance/float(num_node)
def execute(filePath):
############ run experiment ######################
# a = ExperimentExecutor()
# a.autoRun('high_quality_tree/', 'groundTruth.csv', 'compareResult_0124.csv')
###################################################
# PMC2323573_pone.0002033.g002.jpg
try:
phyloParser = PhyloParser(clfPath=clfPath)
# folderPath = 'images/'
# folderPath = 'high_quality_tree/'
# fileNameList = getFilesInFolder(folderPath)
# fileNameList.sort()
# with open('image_fileName_list.txt', 'r')as f:
# fileNameList = f.readlines()
# testingFilePath = [join(folderPath, x) for x in fileNameList]
# result = {}
# groundTruth = {}
# a = ExperimentExecutor()
# a.execute(result, groundTruth, testingFilePath)
# for index, fileName in enumerate(fileNameList):
# print index, fileName
# # f.write(fileName + '\n')
# fileNameList = [x.rstrip() for x in fileNameList]
# fileNameList = [folderPath + x for x in fileNameList]
# for index in range(15, len(fileNameList)):
# try:
# print index, fileNameList[index]
# filePath = folderPath + fileNameList[index]
# fileName = fileNameList[index]
# # fileName = 'PMC2614190_ebo-04-181-g01.jpg'
# # fileName = 'PMC2674049_1471-2148-9-74-4.jpg'
# # fileName = 'PMC1326215_1471-2148-5-71-6.jpg'
# # fileName = 'PMC2644698_1471-2229-8-133-4.jpg'
# # fileName = 'PMC2644698_1471-2229-8-133-4.jpg'
# # fileName = 'PMC2775678_IJMB2009-701735.001.jpg'
# # fileName = 'PMC2697986_1471-2148-9-107-3.jpg'
# # fileName = 'PMC1182362_1471-2148-5-38-3.jpg'
# filePath = folderPath + fileName
# filePath = 'treeset/images/multi/1471-2148-10-52-2-l.jpg'
if isfile(filePath):
image = cv.imread(filePath, 0)
# PhyloParser.displayImage(image)
# imageData = ImageData(image)
# imageData = PhyloParser.preprocces(imageData, debug = False)
# # imageData = PhyloParser.detectLines__v2(imageData, debug=True)
# imageData = PhyloParser.testing(imageData)
image_data = ImageData(image)
image_data = phyloParser.preprocces(image_data, debug=False)
# PhyloParser.displayImage(image)
image_data = phyloParser.detectLines(image_data, debug=False)
# image_data = phyloPaser.traceTree(image_data, debug = True)
image_data = phyloParser.getCorners(image_data, debug=False)
image_data = phyloParser.makeLinesFromCorner(image_data, debug=False)
image_data = phyloParser.includeLinesFromCorners(image_data)
image_data = phyloParser.postProcessLines(image_data)
# image_data = phyloParser.removeRepeatedLines(image_data)
image_data = phyloParser.groupLines(image_data, debug=False)
image_data = phyloParser.matchLineGroups(image_data, debug=False)
# # image_data = phyloParser.detectCorners(image_data)
# # image_data = phyloParser.refineLinesByCorners(image_data)
# # image_data = phyloPaser.getSpecies_v2(image_data, debug = True)
# # image_data = phyloPaser.getSpecies(image_data, debug = True)
# image_data = phyloParser.matchLines(image_data, debug = True, useNew = False)
# image_data = phyloParser.getSpecies_v2(image_data, debug = True)
# # print image_data.orphanBox2Text
image_data = phyloParser.getSpecies_v3(image_data, debug=False)
# testString = phyloParser.makeTree(image_data, debug = True, tracing = False)
# treeString = phyloParser.constructTreeByTracing(image_data, debug = False)
treeString = phyloParser.constructTree_eval(image_data,
debug=False,
tracing=False)
dbdata = treeString.treeHead.getParentChildrenStyle()
# print dbdata['cluster']
with open('dbdata.csv', 'ab') as f:
csvwriter = csv.writer(f, delimiter='\t')
for i in range(dbdata['relation_index'] + 1):
if i >= dbdata['count_labels']:
string = ''
for node in dbdata[i]:
string += '#' + str(node)
csvwriter.writerow([fileName, str(i), string])
else:
csvwriter.writerow([fileName, str(i), dbdata[i]])
with open('dbdata_cluster.csv', 'ab') as f:
csvwriter = csv.writer(f, delimiter='\t')
if 'root' in dbdata['cluster']:
csvwriter.writerow([
fileName,
str(dbdata['relation_index']), dbdata['cluster']['root']
])
for i in range(dbdata['relation_index']):
csvwriter.writerow([fileName, str(i), dbdata['cluster'][i]])
except:
with open('error.txt', 'ab') as f:
f.write(filePath + '\n')