-
Notifications
You must be signed in to change notification settings - Fork 2
/
penerjemah.py
199 lines (164 loc) · 7.23 KB
/
penerjemah.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
from penerjemah_manga.assets.ctd import model2annotations, model2annotation
from ocr.ocr import MangaOcr
from PIL import Image
from tqdm import tqdm
import os
import cv2
import PIL.Image
import json
import numpy as np
mocr = MangaOcr()
# Pengganti cv2_imshow di colab, klo mengunakan colab di hapus saja def ini
def cv2_imshow(img_file):
# Tampilkan gambar menggunakan cv2.imshow
cv2.imshow('Gambar', img_file)
# Tunggu tombol keyboard ditekan dan tutup jendela setelah itu
cv2.waitKey(0)
cv2.destroyAllWindows()
# Definisikan fungsi untuk menggambar kotak pembatas
def draw_bounding_boxes(image, lines):
cropped_images = []
for line in lines:
values = line.split(' ')
x, y, width, height = map(float, values[1:])
width_pixel = int(width * (image.shape[1])) + 20
height_pixel = int(height * image.shape[0]) + 10
x_pixel = int(x * (image.shape[1])) - int((width_pixel - 20) / 2) - 5
y_pixel = int(y * image.shape[0]) - int((height_pixel - 10) / 2) - 10
cv2.rectangle(image, (x_pixel, y_pixel), (x_pixel + width_pixel, y_pixel + height_pixel), (0, 255, 0), 2)
cropped_image = image[y_pixel:y_pixel + height_pixel, x_pixel:x_pixel + width_pixel]
cropped_images.append(cropped_image)
return cropped_images
def draw_white_boxes(image, lines):
for line in lines:
values = line.split(' ')
x, y, width, height = map(float, values[1:])
width_pixel = int(width * (image.shape[1])) + 5
height_pixel = int(height * image.shape[0])
x_pixel = int(x * (image.shape[1])) - int(width_pixel / 2)
y_pixel = int(y * image.shape[0]) - int(height_pixel / 2)
cv2.rectangle(image, (x_pixel, y_pixel), (x_pixel + width_pixel, y_pixel + height_pixel), (255, 255, 255), -1)
def process_image(lokasi_image, lines):
image_polos = cv2.imread(lokasi_image)
draw_white_boxes(image_polos, lines)
return image_polos
# Fungsi untuk menambahkan teks dengan memecah baris jika terlalu panjang
def add_text_multiline(image, text, x, y, max_width, max_height):
font = cv2.FONT_HERSHEY_SIMPLEX
font_scale = 1
font_thickness = 2
font_color = (0, 0, 0) # Warna teks (hitam)
spas = 4
# Fungsi untuk menentukan ukuran font yang sesuai
def get_font_scale(text, max_width, max_height, font=cv2.FONT_HERSHEY_SIMPLEX, font_thickness=1):
font_scale = 1
while True:
(text_w, text_h), _ = cv2.getTextSize(text, font, font_scale, font_thickness)
if text_w <= max_width and text_h <= max_height:
break
font_scale -= 0.1
return font_scale
# Membagi teks menjadi baris-baris yang sesuai dengan lebar maksimum
words = text.split()
tlines = ['']
current_line = 0
for word in words:
test_line = tlines[current_line] + ' ' + word if tlines[current_line] else word
(text_w, text_h), _ = cv2.getTextSize(test_line, font, font_scale, font_thickness)
if text_w <= max_width:
tlines[current_line] = test_line
else:
current_line += 1
tlines.append(word)
# Hitung tinggi total teks
total_text_height = len(tlines) * (text_h + 4) # Spasi antar baris: 2 piksel
if total_text_height>max_height:
font_scale -= 0.5
font_thickness = 1
spas=-1
else:
font_scale = 1
font_thickness = 2
spas=4
# Tentukan posisi vertikal tengah
y_centered = y + int((max_height - total_text_height) / 2) + int(text_h/2) +1
# Menambahkan teks ke gambar
for i, line in enumerate(tlines):
text_size = cv2.getTextSize(line, font, font_scale, font_thickness)[0]
text_x = int(x + (max_width - text_size[0]) / 2)
text_y = y_centered + i * (text_h + spas) # Spasi antar baris: 2 piksel
cv2.putText(image, line, (text_x, text_y), font, font_scale, font_color, font_thickness, lineType=cv2.LINE_AA)
def process_and_add_text(image, lines, tjepang):
# Di dalam loop:
for idx, line in enumerate(lines):
values = line.split(' ')
x, y, width, height = map(float, values[1:])
width_pixel = int(width * (image.shape[1])) + 5
height_pixel = int(height * image.shape[0]) + 10
x_pixel = int(x * (image.shape[1])) - int(width_pixel / 2)
y_pixel = int(y * image.shape[0]) - int(height_pixel / 2)
text = tjepang[idx]
if text!="":
cv2.rectangle(image, (x_pixel, y_pixel), (x_pixel + width_pixel, y_pixel + height_pixel), (255, 255, 255), -1)
add_text_multiline(image, text, x_pixel, y_pixel, width_pixel, height_pixel)
return image
def save_and_translate(img_file, save_dir='hasil', tl_dir='terjemahan'):
if not os.path.exists(save_dir):
os.makedirs(save_dir)
# Mulai Menerjemahkan
model_path = r'penerjemah_manga/assets/comictextdetector.pt'
kordinat = model2annotation(model_path, img_file, save_dir, save_json=False)
if kordinat != ['']:
lines = kordinat[0].split('\n')
#kordinat = kordinat[::-1]
# memotong gambar
image_kotakin = cv2.imread(img_file)
cropped_images = draw_bounding_boxes(image_kotakin, lines)
# Mengambil teks Jepang dari potongan gambar
tjepang = {}
for i, cropped_image in enumerate(tqdm(cropped_images, desc='Processing images')):
cropped_image2 = np.array(cropped_image, dtype=np.uint8)
if cropped_image2 is not None and cropped_image2.any():
img = Image.fromarray(cropped_image2)
text = mocr(img)
else:
text = ""
tjepang[i] = text
#print(f"{i}. '{text}'")
return tjepang,lines
else:
tjepang=""
lines=""
return tjepang,lines
def plance_input_text(img_file, tjepang, lines, save_dir='hasil', tl_dir='terjemahan'):
if tjepang!="":
print("--------------------------------")
print("Terjemahkan json berikut ini:")
print(tjepang)
j1=len(tjepang)
print(f"------------------------------")
tjepang_input = input()
tjepang = eval(tjepang_input)
j2=len(tjepang)
file_name = os.path.basename(img_file)
#print(f"j1={j1} dan j2={j2}")
if (j1==j2):
image = cv2.imread(img_file)
image = process_and_add_text(image, lines, tjepang)
cv2.imwrite(os.path.join(tl_dir, file_name), image)
print(f"---disimpan ke {file_name}")
else:
print(f"Jumlah array kurang! untuk {file_name}")
def plance_text(img_file, tjepang_input, lines, save_dir='hasil', tl_dir='terjemahan'):
file_name = os.path.basename(img_file)
if tjepang_input!="":
tjepang = eval(tjepang_input)
if len(tjepang) == len(lines):
image = cv2.imread(img_file)
image = process_and_add_text(image, lines, tjepang)
cv2.imwrite(os.path.join(tl_dir, file_name), image)
print(f"---disimpan ke {file_name}")
else:
print(f"---Jumlah kalimat tidak sesuai untuk {file_name}")
else:
print(f"Input kosong! atau jumlah text kurang untuk {file_name}")