Freelancer: ManhDuy
Kongsi:
Laporkan Entri

Yolov5 and OCR

Dear client, This is my entry, yolov5 + OCR I downloaded 160 images from your web to train a detection model. with it i can know the position of two values. If there are more photos, I can do better. Once I got the position of the 2 values, I cut them out of the image and put in the OCR (Optical Character Recognition) model and I got the result. The whole model is very light, you can run on embedded or mobile devices. I can also tweak it so it can work with more precision on expensive devices. Please take a look at some test results Best regards, thanks.


                                                                                                            Penyertaan Peraduan #                                        6
                                     untuk                                         Retrieve numbers from graphic images using python cv2 and tesseract

Papan Penjelasan Umum

  • karmafunk
    Penganjur Peraduan
    • 2 bulan yang lalu

    Great, then can we integrate your code with our codes? Do you want to see our code?

    • 2 bulan yang lalu
  • ManhDuy
    ManhDuy
    • 3 bulan yang lalu

    with deep learning, you get a lot more accuracy than just using cv2.

    • 3 bulan yang lalu
  • ManhDuy
    ManhDuy
    • 3 bulan yang lalu

    yep, i used python, cv2 and deep learning.

    • 3 bulan yang lalu
  • karmafunk
    Penganjur Peraduan
    • 3 bulan yang lalu

    Did you use python and cv2?

    • 3 bulan yang lalu