Search code examples
python-3.xpython-tesseractpytesserimg2pdf

How to convert .png images to text using Python


Recently, I took a project. Converting a scanned PDF to searchable PDF/word using Python tesseract.

After few attempts, I could able to convert scanned PDF to PNG image files and afterwards, I'm struck could anyone please help me to convert the PNG files to Word/PDF searchable.my piece of code attached

Please find the attached image for reference.

Import os
Import sys
from PIL import image
Import pytesseract
from pytesseract import image_to_string

 Libpath =r'_______' #site-package
 Pop_path=r'_______' #poppler dlls
 Sys.path.insert(0,LibPath)

  from pdf2image import convert_from_path

     Pdfpath=r'_______' # PDF file directory
     imgpath=r'_______' #image output path

     images= convert_from_path(pdf_path = pdfpath, 
         dpi=500, poppler_path= pop_path)
      for idx, of in enumerate (images):
                 pg.save(imgPath+'PDF_Page_'+'.png',"PNG")
                 print('{} page converted'.format(str(idx)))

       try:
          from PIL import image
       except ImportError:
                 import image
         import pytesseract

     def ocr-core(images):
              Text = 
       pytesseract.image_to_string(image.open(images))
       return text
  print(ocr_core("image path/imagename))

that's it, I've written.....then I got multiple ".PNG" images...now I can only able to convert one PNG images to text.

How to convert all the images and save it in CSV/word?


Solution

  • I got answer to my question

    from PIL import Image
    from pdf2image import convert_from_path
    import pytesseract
    
    
    Pdf_file_path = '_______' #your file path
    
    Images = convert_from_path(Pdf_file_path, dpi=500)
    
    Counter=1
    for page in Images:
           idx= "image_"+str(Counter)+".jpg" ##or ".png"
           page.save(idx, 'JPEG')
           Counter += 1
    
     file=Counter-1
     output= '_____' #where you want to save and file name
     f=open(output, "w")
     for i in range(1,file+1):
              idx= "image_"+str(Counter)+".jpg" ##or ".png"         
              text=str(pytesseract.image_to_string(Image.open(idx)))
              f.write(text)
     f.close()