As a tool of information transmission, QR code has played an important role in today's society. From mobile phone users logging in to mobile payment, QR codes can be seen in every corner of life. Do you know how the QR code is parsed? Have you ever thought about implementing a code scanning tool yourself? Keep reading if you want!
Let us first think about what operations need to be written to realize the scanning tool. In the process of scanning the code, we need to open the camera, how to recognize the QR code from the mobile phone or computer. So we have to implement two key steps: call the camera and identify the QR code.
These two operations correspond to two modules respectively, they are opencv and pyzbar, where opencv is Intel's computer vision processing module, and pyzbar is a module used to parse QR codes.
The environment includes python environment and modules. My environment is as follows:
System: Windows 10
python：python 3.7.9
opencv：opencv-python-4.4.0.44
pyzbar：pyzbar-0.1.8
Module installation is very simple, we install directly with pip, first install the opencv module:
pip install opencv-python
Then the pyzbar module:
pip install pyzbar
When the installation version is not specified, the system will automatically install the latest version. After installing the module, we can implement the code scanning tool.
With the pyzbar module, our job of identifying the QR code is very simple. First, we need to prepare a QR code. Once you have the QR code, you can start parsing. The specific steps are as follows:
The implementation code is as follows:
import cv2
from pyzbar import pyzbar
# 1、 Read QR code image
qrcode = cv2.imread('qrcode.jpg')
# 2、 Analyze the data in the QR code
data = pyzbar.decode(qrcode)print(data)
# 3、 Analyze the data information of the QR code in the data
text = data[0].data.decode('utf-8')print(text)
We parsed the above twice, and got a data for the first time. Let's take a look at what the data looks like:
[ Decoded(data=b'http://weixin.qq.com/r/vC_fhynEKnRVrW3k93qu', type='QRCODE', rect=Rect(left=140, top=113, width=390, height=390), polygon=[Point(x=140, y=113),Point(x=140, y=503),Point(x=530, y=503),Point(x=530, y=113)])]
You can see that it is a list, and the first data of the list contains url information. So we need to parse again through the following code:
text = data[0].data.decode('utf-8')
So we can get the information contained in the QR code. In order to facilitate subsequent use, the above code can be written as a function:
def scan_qrcode(img_path):
qrcode = cv2.imread(img_path)
data = pyzbar.decode(qrcode)return data[0].data.decode('utf-8')
Next we will look at how to call the camera.
A VideoCapture class is provided in opencv to read the video, which can also be used to call the camera. The steps to call the camera are as follows:
The specific code is as follows:
import cv2
# Call camera
cap = cv2.VideoCapture(0)while True:
# Read a frame
ret, frame = cap.read()
# Show current frame
cv2.imshow('scan qrcode', frame)
# Waiting for keyboard input
key = cv2.waitKey(10)
# Turn off the camera when the q key is pressed
if key ==ord('q'):break
# Destroy all windows
cv2.destroyAllWindows()
You can try to run the above code yourself, the effect is like turning on your front camera.
Now that the camera is called, we can combine the two parts of the code.
The main part of our code scanning tool is the operation of calling the camera. We need to parse each frame we read, and output and exit when the result is parsed. The specific code is as follows:
import cv2
from pyzbar import pyzbar
def scan_qrcode(qrcode):
data = pyzbar.decode(qrcode)return data[0].data.decode('utf-8')
cap = cv2.VideoCapture(0)while True:
ret, frame = cap.read()
cv2.imshow('scan qrcode', frame)
# Analyze the QR code
text = None
try:
text =scan_qrcode(frame)
except Exception as e:
pass
if text:print(text)break
key = cv2.waitKey(10)if key ==ord('q'):break
cv2.destroyAllWindows()
We modified the scan_qrcode function above, from the original path of the incoming image to the direct incoming image object. Because the picture frame obtained through the VideoCapture object and the picture obtained through cv2.imread are of the same data type.
The key step above is to analyze the operation of the QR code. First define a text, because an exception will occur if there is no QR code during the parsing process, so use the try-except statement to handle it. How to judge the content of text by if, only when we really parse the data, the program will output the result and exit the program.
At this point, we have realized the scanning tool.
Recommended Posts