Question

I am using the PIL (python image library) to crop a very large image and present the cropped area to the interface. The problem Im having is that the process is taking too long. When the user clicks on the image to crop it, the image takes quite a long time to show up on the sizer I attach it to.

I tried doing this two ways: First I tried saving the cropped area as an image to the disk, and loaded it on the fly into the sizer. The second attempt was to create an empty image and convert the pil image into the wx image and load that onto the sizer. Surprising to me is that the first method of writing to the disk feels faster than the second method of managing it in memory. Here are the code samples:

First method:

area = image_object.crop(self.cropxy)
area.save(CROP_IMAGE, 'jpeg')
crop_image  = wx.Image(CROP_IMAGE, wx.BITMAP_TYPE_JPEG).ConvertToBitmap()
crop_bitmap = wx.StaticBitmap(self.crop_panel, bitmap=crop_image, name="Cropped Image")
crop_bitmap.CenterOnParent()
crop_bitmap.Refresh()

Second method:

area = image_object.crop(self.cropxy)
image = wx.EmptyImage(area.size[0], area.size[1])
image.SetData(area.convert("RGB").tostring())
crop_image = wx.BitmapFromImage(image)
crop_bitmap = wx.StaticBitmap(self.crop_panel, bitmap=crop_image, name="Cropped Image")
crop_bitmap.CenterOnParent()
crop_bitmap.Refresh()

Is there a better way to do this so that the image will now show up so slowly?

Answer 1

So in order to solve something somewhere else in the interface, when I queue up my images I decided to pre-load the wxImage objects. Never had to before when they were much smaller.

Anyway – I found some code on google that would allow me to convert between wxImage objects and PIL objects and by doing so, I can convert the in-memory wxImage object to the PIL object, crop it, and convert it back to the image just in time to display it. This is ‘Blazing’ fast by comparison. You just hardly take your finger off the mouse and the crop shows just fine.

Here are the conversion routines:

def pil_to_image(self, pil, alpha=True):
    """ Method will convert PIL Image to wx.Image """
    if alpha:
        image = apply( wx.EmptyImage, pil.size )
        image.SetData( pil.convert( "RGB").tostring() )
        image.SetAlphaData(pil.convert("RGBA").tostring()[3::4])
    else:
        image = wx.EmptyImage(pil.size[0], pil.size[1])
        new_image = pil.convert('RGB')
        data = new_image.tostring()
        image.SetData(data)
    return image


def image_to_pil(self, image):
    """ Method will convert wx.Image to PIL Image """
    pil = Image.new('RGB', (image.GetWidth(), image.GetHeight()))
    pil.fromstring(image.GetData())
    return pil