-1

i would like to know if easyocr for license plate recognition is dependant on the GPU. Im trying to use this script to read the license plate in a image and it's EXTREMELY SLOW (10+ min). And i just found out the computer i'm using does not have a GPU.

 from PIL import Image
    import numpy as np
    import easyocr
    import ssl
    ssl._create_default_https_context = ssl._create_unverified_context
    
    IMAGE_PATH = 'test.png'
    reader = easyocr.Reader(['en'])
    result = reader.readtext(IMAGE_PATH)
    for detection in result:
        if detection[2] > 0.5:
            print(detection[1])
Master
  • 135
  • 1
  • 11

2 Answers2

4

As seen on the pypi installation page of easyocr

"In case you do not have a GPU, or your GPU has low memory, you can run the model in CPU-only mode by adding gpu=False.

reader = easyocr.Reader(['ch_sim','en'], gpu=False)

"

Anonymous
  • 452
  • 2
  • 9
  • Thanks for the help. Yes,i tried doing so, it works but it's still extremely slow. I guess if i use a better pc (CPU or GPU) it will probably work better, right? – Master Jul 28 '22 at 20:35
  • 1
    @Master yeah, lots of machine learning/data processing stuff is better suited for the GPU, sometimes thousands of times faster than the CPU, depending on the task at hand – Anonymous Jul 28 '22 at 20:38
-2

EasyOCR works without GPU, but is crushingly slow.

andrewJames
  • 19,570
  • 8
  • 19
  • 51
Colim
  • 1,283
  • 3
  • 11