def perform_ocr(img_url):

    #use dictionary or not
    use_dict = True

    raw_image = cv2.imread(img_url, 0)
    #cv2.imshow('image',raw_image)
    #cv2.waitKey(0)

    #get all the words (as an numpy image array), words on each line, and maximum height on that line
    all_words, words_on_line, max_height_on_line = get_words(raw_image)
    #print('ocr: ',all_words)

    print("Total no. of lines = ", len(words_on_line))
    print("Words per line = ", words_on_line)
    print("Total no. of words = ", len(all_words))

    #to write the output into a file
    fp = open("output.txt", 'w')
    fp.truncate()

    #start loading bar
    bar = Bar('Loading...', max=len(all_words))

    count = 0
    for i in range(0, len(words_on_line)):

        for j in range(0, words_on_line[i]):

            #update loading bar
            bar.next()
            all_characters = get_characters(all_words[count],
                                            max_height_on_line[i], i, j)

            if use_dict:
                # print (correction(get_string_from_nn(all_characters)),)
                fp.write(correction(get_string_from_nn(all_characters)))
                fp.write(" ")
            else:
                # print (get_string_from_nn(all_characters),)
                fp.write(get_string_from_nn(all_characters))
                fp.write(" ")

            # exit(0)
            # cv2.imshow("all_words[count]",all_words[count])
            # cv2.waitKey()

            count = count + 1

        #print ("\n")
        fp.write("\n")
        # bar.next()
    #close loading bar
    bar.finish()

    fp.close()
    print("\nconversion completed successfully")
def extract_character(words_on_line,
                      all_words,
                      max_height_on_line,
                      use_dict=False):
    #to write the output into a file
    fp = open("output.txt", 'w')
    fp.truncate()
    count = 0
    for i in range(0, len(words_on_line)):
        for j in range(0, words_on_line[i]):
            all_characters = cp.get_characters(all_words[count],
                                               max_height_on_line[i], i, j)
            if use_dict:
                None
                # print (correction(get_string_from_nn(all_characters)))
                # fp.write(correction(get_string_from_nn(all_characters)))
                # fp.write(" ")
            else:
                word = user_input.get_string_from_nn(all_characters)
                print(str(count) + " word - " + word)
                fp.write(word)
                fp.write(" ")
            # exit(0)
            # cv2.imshow("all_words[count]",all_words[count])
            # cv2.waitKey()
            count = count + 1
        # print ("\n")
        fp.write("\n")
    fp.close()
    return
def perform_ocr(img_url):

    #use dictionary or not
    use_dict = True

    raw_image = cv2.imread(img_url, 0)
    #cv2.imshow('image',raw_image)

    #get all the words (as an numpy image array), words on each line, and maximum height on that line
    all_words, words_on_line, max_height_on_line = get_words(raw_image)
    #print('ocr: ',all_words)

    print("no. of lines = ", len(words_on_line))
    print(words_on_line)
    print("no. of words = ", len(all_words))

    #to write the output into a file
    fp = open("output.txt", 'w')
    fp.truncate()

    count = 0
    for i in range(0, len(words_on_line)):

        for j in range(0, words_on_line[i]):

            all_characters = get_characters(all_words[count],
                                            max_height_on_line[i], i, j)

            if use_dict:
                print(correction(get_string_from_nn(all_characters)), )
                fp.write(correction(get_string_from_nn(all_characters)))
                fp.write(" ")
            else:
                print(get_string_from_nn(all_characters), )
                fp.write(get_string_from_nn(all_characters))
                fp.write(" ")

            # exit(0)
            # cv2.imshow("all_words[count]",all_words[count])
            # cv2.waitKey()

            count = count + 1

        #print ("\n")
        fp.write("\n")

    fp.close()
Esempio n. 4
0
File: ocr.py Progetto: fanzalika/ocr
def perform_ocr(img_url):
	
	#use dictionary or not
	use_dict = True
	
	raw_image = cv2.imread(img_url,0)

	#get all the words (as an numpy image array), words on each line, and maximum height on that line
	all_words, words_on_line, max_height_on_line = get_words(raw_image)

	print "no. of lines = ",len(words_on_line)
	print words_on_line
	print "no. of words = ",len(all_words)

	#to write the output into a file
	fp = open("output.txt", 'w')
	fp.truncate()

	count = 0
	for i in range(0, len(words_on_line)):

		for j in range(0, words_on_line[i]):
			
			all_characters = get_characters(all_words[count],max_height_on_line[i],i,j)
			
			if use_dict:
				print correction(get_string_from_nn(all_characters)),
				fp.write(correction(get_string_from_nn(all_characters)))
				fp.write(" ")
			else:
				print get_string_from_nn(all_characters),
				fp.write(get_string_from_nn(all_characters))
				fp.write(" ")
				
			# exit(0)
			# cv2.imshow("all_words[count]",all_words[count])
			# cv2.waitKey()
			
			count = count + 1
			
		print "\n"
		fp.write("\n")

	fp.close()
Esempio n. 5
0
def perform_ocr(img_url):

    raw_image = cv2.imread(img_url, 0)

    #get all the words (as an numpy image array), words on each line, and maximum height on that line
    all_words, words_on_line, max_height_on_line = get_words(raw_image)

    print "no. of lines = ", len(words_on_line)
    print words_on_line
    print "no. of words = ", len(all_words)

    #to write the output into a file
    fp = open("output.txt", 'w')
    fp.truncate()

    count = 0
    for i in range(0, len(words_on_line)):

        for j in range(0, words_on_line[i]):

            all_characters = get_characters(all_words[count],
                                            max_height_on_line[i], i, j)

            print get_string_from_nn(all_characters),
            fp.write(get_string_from_nn(all_characters))
            fp.write(" ")

            # exit(0)
            # cv2.imshow("all_words[count]",all_words[count])
            # cv2.waitKey()

            count = count + 1

        print "\n"
        fp.write("\n")

    fp.close()