Separate/reposition/translate shapes in image with pillow in python

问题

I need to separate or translate or replace pixels in an image with python so as to all the shapes to share the same distance between each other and the limits of the canvas.

background is white, shapes are black.

IMPORTANT: Images are dynamic, all images have bars at different positions, that means i need to detect where a bar starts and where a bar ends to draw the final image!

This is an example INPUT image

This is an example OUTPUT image

I did this by hand, i don't know if all bars are separated by the same distance, it's just an example.

The resulting canvas size can change too but ideally i would like all canvas to have the same width and height after the bars have been respositioned.

All pixels in the images are BLACK or WHITE, that should make it much easier.

What i've tried so far kind of detects the shapes but some shapes are not detected correctly when they are too close of each other:

saut = 1
start_ab = 0
end_ab = 0
start_or = im2.size[1] - 1
end_or = 0
size = 0

shapes = []

for y in range(0, im2.size[0], saut):  # slice across
    for x in range(0, im2.size[1], 1):  # slice down
        pix = im2.getpixel((y, x))
        if pix != 255:
            start_or = min(start_or, x)
            end_or = max(end_or, x)
            inshape = True

    if foundshape == False and inshape == True:
        foundshape = True
        start_ab = y

    if foundshape == True and inshape == False:
        foundshape = False
        end_ab = y
        size = max(size, end_ab)
        shapes.append((start_ab, end_ab, start_or, end_or))
        start_or = im2.size[1] - 1
        end_or = 0
    inshape = False

print shapes
# example shapes output NOT FROM THE EXAMPLE IMAGES PROVIDED but from other shapes [(54, 171, 72, 233), (216, 324, 108, 251), (342, 486, 0, 215), (513, 765, 18, 260), (792, 918, 90, 242)]

And still have to draw the new image, i don't know how to do that, also some images have shapes with "holes" in them, so that makes the "redraw" a little bit more complicated.

回答1:

And still have to draw the new image, i don't know how to do that, also some images have shapes with "holes" in them, so that makes the "redraw" a little bit more complicated.

Once you've established the bounds of a rectangle, you could use the Image.crop() function to extract a rectangle from the image (including those holes) and Image.paste() to paste them in the right position. The tutorial contains an example:

http://pillow.readthedocs.org/en/latest/handbook/tutorial.html#cutting-pasting-and-merging-images

回答2:

Code to calculate the distance between each shape. It assumes that the shapes have all the same width:

def calcDistanceBetweenShapes(nShapes, shapeWidth, totalWidth):
    totalBlank = totalWidth-nShapes*shapeWidth
    return totalBlank/nShapes

Code to shift a shape's abscissa (assumes a shape is stored as (start_ab, end_ab, start_or, end_or,empty_points)):

def setNewAbscissa(shape, new_start_ab):
    start_ab, end_ab = shape[0], shape[1]
    shift = new_start_ab - start_ab
    new_end_ab = end_ab + shift
    new_empty_points = map(lambda pt: (pt[0]+shift,pt[1]),empty_points)
    return (new_start_ab, new_end_ab,start_or, end_or, new_empty_points)

Now let's call d the distance between shapes. The first shape will not be moved, so its coordinates stay the same. The second shape will have its start_ab that will be previous_end_ab+d where previous_end_ab is the end_ab of the first shape, etc. This gives an algorithm to calculate the coordinates of all the shapes.

def get_new_shapes(shapes, totalWidth):
    shift = calcDistanceBetweenShapes(len(shapes), shapes[0][1] - shape[0][0], totalWidth)
    res = [shapes[0]]
    previous_shape = shapes[0]
    for shape in shape[1:]:
         new_start_ab = previous_shape[1] + shift
         new_shape =  setNewAbscissa(previous_shape,new_start_ab)
         res.append(new_shape)
         previous_shape = new_shape

来源：https://stackoverflow.com/questions/31966970/separate-reposition-translate-shapes-in-image-with-pillow-in-python