CODEWITHSUNDEEP
pytorchqr-codeobject-detection
Y W
Y W

Reputation: 25

which object detection algorithms can extract the QR code from an image efficiently

I am new to Object Detection, for now, I want to predict the QR code in images. I want to extract the QR code from the images, and only predict the QR code without the background information, and finally predict the exact number the QR code is representing, since I am using PyTorch, is there any object detection algorithm which is compatible to PyTorch that I could apply to this task? (for what I mean by extracting, the raw input is an image,I want to change the input of the image to the QR code in the image).

Upvotes: 0

Views: 1129

Answers (1)

yakhyo
yakhyo

Reputation: 1656

There are two ways for this task:

Computer Vision based approach:

OpenCV library's QRCodeDetector() function can detect and read QR codes easily. It returns data in QR code and bounding box information of the QR code:

import cv2
detector = cv2.QRCodeDetector()
data, bbox, _ = detector.detectAndDecode(img)

Deep learning based approach:

Using common object detection framework - Yolo (Yolo v5 in PyTorch), you can achieve your target. However, you need data to train it. For computer vision based approach, you don't need to do training or data collection.

You may consider reading these two.

Upvotes: 1

Related Questions