def processLabeledImages(self):
image = self.image_processing_list[0]
boundries = ImageProcessing.GetLineBounds(image.imageArray)
line_images_array = []
for i in range(len(boundries)):
x, y, w, h = boundries[i]
cutImage = image.cutImage(image.imageArray, x, y, x + w, y + h)
line_images_array.append(cutImage)
Main.userSetLabel(line_images_array, self)
def extract_ines_of_text_from_image_and_match_label(imageArray, text,fontname, folderSave):
lines_txt = text.split('\n')
num_lines_txt = len(lines_txt)
print(num_lines_txt)
num_line=0
newImagesForTrain = []
width = imageArray.shape[1]
height = imageArray.shape[0]
boundries = ImageProcessing.GetLineBounds(imageArray)
print(len(boundries))
num_lines_image = len(boundries)
for i in range(len(boundries)):
x, y, w, h = boundries[i]
cutImage = imageArray[max(0, min(y, y+h)-NUMPIXELS):min(max(y, y+h)+NUMPIXELS,height) , max(0, min(x, x+w)-NUMPIXELS) :min(width, max(x, x+w)+NUMPIXELS)]
while lines_txt[num_line]=="" or lines_txt[num_line]==" " and num_line<num_lines_txt-1:
num_line+=1
if num_line<=num_lines_txt-1:
Label = lines_txt[num_line]
num_line += 1
newImagesForTrain.append(ImageProcessing.ImageProcessing(cutImage, imagePath=None, handwrite_ID=fontname, Label = Label))
Insert_to_folder(newImagesForTrain, folderSave)
def extract_ines_of_text_from_images_and_match_labels(folder, txtfile):
font_name = os.path.basename(folder)[:-8] #remove "docx_" and "_images_"
font_origin = font_name
if len(font_name)>5:
font_name = font_name[0:10]
while font_name in names_of_fonts:
font_name= font_name+"a"
names_of_fonts.append(font_name)
files = os.listdir(folder)
if font_name in files:
#shutil.rmtree(os.path.join(folder, font_name))
return
os.mkdir(os.path.join(folder, font_name))
folderSave = os.path.join(folder, font_name)
till_page = len(files)
print(till_page )
txt_file = open(txtfile, "r" , encoding="utf-8")
text = txt_file.read()
txt_file.close()
lines_txt = text.split('\n')
num_lines_txt = len(lines_txt)
print(num_lines_txt)
num_line=0
newImagesForTrain = []
num_lines_images = 0
for i in range(till_page-1):
print(i)
image = os.path.join(folder, font_origin+"_"+str(i)+".tif")
print(image)
imageArray = cv.imread(image, 0)
width = imageArray.shape[1]
height = imageArray.shape[0]
boundries = ImageProcessing.GetLineBounds(imageArray)
#print(boundries)
print(len(boundries))
thispageSnum = num_lines_images
num_lines_images += len(boundries)
for i in range(len(boundries)):
x, y, w, h = boundries[i]
cutImage = imageArray[max(0, min(y, y+h)-NUMPIXELS):min(max(y, y+h)+NUMPIXELS,height) , max(0, min(x, x+w)-NUMPIXELS) :min(width, max(x, x+w)+NUMPIXELS)]
while lines_txt[num_line]=="" or lines_txt[num_line]==" " and num_line<num_lines_txt-1:
num_line+=1
if num_line<=num_lines_txt-1:
Label = lines_txt[num_line]
print(Label)
#cv.imshow("bb", cutImage)
#cv.waitKey(0)
num_line += 1
print(thispageSnum+i)
print(num_line)
newImagesForTrain.append(ImageProcessing.ImageProcessing(cutImage, imagePath=None, handwrite_ID=font_name, Label = Label))
Insert_to_folder(newImagesForTrain, folderSave)
print(num_line, num_lines_images)
