PyQt5 Convert 2d np array to QImage

Question

from PyQt5.QtWidgets import QApplication, QWidget, QLabel, QVBoxLayout
from PyQt5.QtGui import QPixmap, QImage
import numpy as np
import sys

def print_np():
    app = QApplication(sys.argv)

    win = QWidget()
    label = QLabel()

    img = np.ones((500, 500))
    qImg = QPixmap(QImage(img.data, img.shape[0], img.shape[1], QImage.Format_Indexed8))
    label.setPixmap(qImg)

    vbox = QVBoxLayout()
    vbox.addWidget(label)
    win.setLayout(vbox)
    win.show()

    sys.exit(app.exec_())

In this code, I am trying to display 2d np array on QtWindow.

I am expecting to see a white square, but I keep getting a weird display as below.

I believe its because wrong QImage.Format (data type), so I tried every option but didn't help.

https://srinikom.github.io/pyside-docs/PySide/QtGui/QImage.html#PySide.QtGui.QImage

Should I change np array to jpg or png first, as examples in documentation?

Why have you removed the #import statements? It makes your code impossible to run and it's therefore harder to assist you. — Mark Setchell
– Mark Setchell, Commented Feb 13, 2020 at 8:53
Are you hoping to display colours in your image? If so, it needs to be of shape 500,500,3 to accommodate 3 RGB channels and have type RGB888. If you only want black/white/greys, your shape is correct but you need to tell Qt is is greyscale. — Mark Setchell
– Mark Setchell, Commented Feb 13, 2020 at 9:06
Your array is float64, when you needed int8: img = np.ones((500, 500), dtype=dtype=np.uint8). And maybe you also want the RGB version img = np.ones((3, 500, 500) ... — Jean-Didier
– Jean-Didier, Commented Feb 13, 2020 at 9:10
` img = np.zeros((500, 500, 3), dtype=np.uint8) qImg = QPixmap( QImage(img.data, img.shape[0], img.shape[1], QImage.Format_RGB888)) label.setPixmap(qImg) ` Both np.zeros and ones give me a black image. Could you explain why? — Sean
– Sean, Commented Feb 13, 2020 at 9:33

Sean · Accepted Answer · 2020-02-13 18:26:13Z

4

There are a couple of issues:

Your dtype needs to be np.uint8 when you create your Numpy image array
Your format needs to be QImage.Format_RGB888
Your pixels need to bigger than 0 or 1 to show up, the full range is 0..255

So, with that in mind:

#!/usr/bin/env python3

from PyQt5.QtWidgets import QApplication, QWidget, QLabel, QVBoxLayout
from PyQt5.QtGui import QPixmap, QImage
import numpy as np
import sys

def main():
    app = QApplication(sys.argv)

    win = QWidget()
    label = QLabel()

    img = np.zeros((500, 500, 3), dtype=np.uint8)
    # Turn up red channel to full scale
    img[...,0] = 255
    qImg = QPixmap(QImage(img.data, img.shape[0], img.shape[1], QImage.Format_RGB888))
    label.setPixmap(qImg)

    vbox = QVBoxLayout()
    vbox.addWidget(label)
    win.setLayout(vbox)
    win.show()

    sys.exit(app.exec_())

main()

++ (to convert 2D image as black vs white)

img = np.zeros((500, 500), dtype=np.uint8)
img[250:, :] = 255

qImg = QPixmap(QImage(img.data, img.shape[0], img.shape[1], QImage.Format_Grayscale8))

It will create a window

Keywords: Python, Qt, QImage, QPixmap, image processing.

edited Feb 13, 2020 at 18:26

Sean

5191 gold badge10 silver badges29 bronze badges

answered Feb 13, 2020 at 12:33

Mark Setchell

210k32 gold badges310 silver badges504 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Mark Setchell Over a year ago

@Sean Rather than QImage.Format_Indexed8, you should probably use QImage.Format_Grayscale8 to be sure of tones.

Sean Over a year ago

I edited once more, btw thank you so much for your answer!

Mark Setchell Over a year ago

Cool, and thanks also to @Demi-Lune Good luck with your project and remember, questions (and answers) are free, so come back if you get stuck.

Collectives™ on Stack Overflow

PyQt5 Convert 2d np array to QImage

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related