Display numpy ndarray in wxpython

Question

I want to display a numpy ndarray in wxpython but somehow I can't get it to work...

My image control:

self.imageCtrl = wx.StaticBitmap(self.panel, wx.ID_ANY, wx.BitmapFromImage(img))

My image code:

img = cv2.imread(file, 0)
flag, thresh = cv2.threshold(img, 100, 255, cv2.THRESH_BINARY
# thresh is an ndarray with size [60, 60]
img_wx = wx.ImageFromBuffer(thresh.shape[0], thresh.shape[1], thresh)
self.imageCtrl.SetBitmap(wx.BitmapFromImage(img_wx)

This results in the following error:

Traceback (most recent call last):
File "cropImagesAndLabel.py", line 114, in onStart
self.countUp()
File "cropImagesAndLabel.py", line 134, in countUp
self.crop()
File "cropImagesAndLabel.py", line 104, in crop
img = wx.ImageFromBuffer(self.pattern.shape[0], self.pattern.shape[1], self.pattern)
File "\Canopy\User\lib\site-packages\wx\_core.py", line 3598, in ImageFromBuffer
image = _core_._ImageFromBuffer(width, height, dataBuffer, alphaBuffer)
TypeError: expected a single-segment buffer object

It would be great if anybody could help!

Kind regards, m

Answer 1

cv2's image is actually a ndarry, you can use Matplotlib window to show the image. You can also put the Matplotlib's window inside a wxPython window, see this as a reference: user_interfaces example code: embedding_in_wx2.py. Note that the link above shows how to display a graph using the plot function. You can use self.axes.imshow(your_image) to show the image. For example, a black image can be constructed by following statement

your_image = np.zeros((512,512,3), 'uint8')

Also, you should note that in the above link, the backend of the matplotlib has set to wxAgg, which means it use wxPython as a drawing backend.