| Foydalanilgan adabiyotlar:
1. Simões, J., & Ribeiro, R. (2017). A Survey on Text Detection and Extraction in Natural Scenes. Journal of Imaging, 3(3), 40. doi:10.3390/jimaging3030040
2. Mishra, A., Alahari, K., & Jawahar, C. V. (2012). Scene Text Recognition using Higher Order Language Priors. In Proceedings of the European Conference on Computer Vision (ECCV) (pp. 548-562). Springer.
3. Wang, K., Babenko, B., & Belongie, S. (2011). End-to-end Scene Text Recognition. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 1457-1464). IEEE.
4. Liu, Y., Jin, L., Zhang, S., Zhang, C., & Lu, X. (2019). Deep Text Recognition in Natural Scene Images: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 41(11), 2592-2606. doi:10.1109/TPAMI.2018.2876851
5. Shi, B., Bai, X., & Yao, C. (2017). An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 39(11), 2298-2304. doi:10.1109/TPAMI.2016.2635203
6. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. In Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) (pp. 234-241). Springer.
7. International Research Journal of Engineering and Technology (IRJET). Text Detection and Recognition: A Review
Foydalanilgan internet saytlar
https://colab.research.google.com/
https://www.mygreatlearning.com/blog/introduction-to-image-pre-processing/
https://www.ibm.com/blog/optical-character-recognition/
https://nanonets.com/blog/ocr-with-tesseract/
https://www.researchgate.net/publication/274838198_A_Technical_Review_on_Text_Recognition_from_Images
Ilova:
# Kerakli kutubxonalarni o'rnatish
!sudo apt install tesseract-ocr
!pip install pytesseract
!pip install opencv-python
from google.colab import files
import re
import cv2
import pytesseract
from PIL import Image
import IPython.display as display
from io import BytesIO
import base64
# Tasvirda OCRni bajarish
def perform_ocr(image_path):
# OpenCV yordamida rasmni o'qish
img = cv2.imread(image_path)
# Tasvirni kulrang rangga aylantirish
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# Pytesseract yordamida OCR ni bajarish
text = pytesseract.image_to_string(Image.fromarray(gray))
# Qo'shimcha bo'sh qatorlarni olib tashlash
lines = text.split('\n')
filtered_lines = []
empty_line_count = 0
for line in lines:
if line.strip() == '':
empty_line_count += 1
if empty_line_count <= 1:
filtered_lines.append(line)
else:
empty_line_count = 0
filtered_lines.append(line)
result_text = '\n'.join(filtered_lines)
return result_text
# Tasvir va matnni yonma-yon ko'rsatish funktsiyasi
def display_image_with_text(image_path, text):
# Ko'rish uchun tasvir hajmini o'zgartirish
img = Image.open(image_path)
img.thumbnail((300, 300)) # O'lchamni sozlash
img_byte_array = BytesIO()
img.save(img_byte_array, format='PNG', quality=95) # Tasvir sifatini yaxshilash
img_data = img_byte_array.getvalue()
# Matn boshlanishidan oldin va keyin 5 qator bo'sh joy qo'shish
text = '\n' * 5 + text.strip() + '\n' * 5
# HTML yordamida rasm va matnni yonma-yon ko'rsatish
display.display(display.HTML(f""))
# Tasvirlarni tsiklda qayta ishlash
while True:
# Rasmni yuklash
uploaded_files = files.upload()
if not uploaded_files:
break # Hech qanday fayl yuklanmaganda tsikldan chiqarish
# Har bir yuklangan rasmni qayta ishlash va namoyish qilish
for image_filename, image_content in uploaded_files.items():
# Rasmni saqlash
with open(image_filename, 'wb') as f:
f.write(image_content)
# Rasmda OCRni bajarish
result_text = perform_ocr(image_filename)
# Rasm va matnni ko'rsatish
display_image_with_text(image_filename, result_text)
# Qayta ishlangan rasm uchun yuklab olish havolasini taqdim etish
download_link = f'Yuklangan {image_filename}'
display.display(display.HTML(download_link))
# Davom etishni so'rash
user_input = input("Yana rasm yuklashni xohlaysizmi? (xa/yo'q): ").lower()
if user_input != 'xa':
break # Agar foydalanuvchi davom etishni istamasa, tsikldan chiqish
|
