crop center portion of a numpy image

Something along these lines -

def crop_center(img,cropx,cropy):    y,x = img.shape    startx = x//2-(cropx//2)    starty = y//2-(cropy//2)        return img[starty:starty+cropy,startx:startx+cropx]

Sample run -

In [45]: imgOut[45]: array([[88, 93, 42, 25, 36, 14, 59, 46, 77, 13, 52, 58],       [43, 47, 40, 48, 23, 74, 12, 33, 58, 93, 87, 87],       [54, 75, 79, 21, 15, 44, 51, 68, 28, 94, 78, 48],       [57, 46, 14, 98, 43, 76, 86, 56, 86, 88, 96, 49],       [52, 83, 13, 18, 40, 33, 11, 87, 38, 74, 23, 88],       [81, 28, 86, 89, 16, 28, 66, 67, 80, 23, 95, 98],       [46, 30, 18, 31, 73, 15, 90, 77, 71, 57, 61, 78],       [33, 58, 20, 11, 80, 25, 96, 80, 27, 40, 66, 92],       [13, 59, 77, 53, 91, 16, 47, 79, 33, 78, 25, 66],       [22, 80, 40, 24, 17, 85, 20, 70, 81, 68, 50, 80]])In [46]: crop_center(img,4,6)Out[46]: array([[15, 44, 51, 68],       [43, 76, 86, 56],       [40, 33, 11, 87],       [16, 28, 66, 67],       [73, 15, 90, 77],       [80, 25, 96, 80]])

A more general solution based on @Divakar 's answer:

def cropND(img, bounding):    start = tuple(map(lambda a, da: a//2-da//2, img.shape, bounding))    end = tuple(map(operator.add, start, bounding))    slices = tuple(map(slice, start, end))    return img[slices]

and if we have an array a

>>> a = np.arange(100).reshape((10,10))array([[ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9],       [10, 11, 12, 13, 14, 15, 16, 17, 18, 19],       [20, 21, 22, 23, 24, 25, 26, 27, 28, 29],       [30, 31, 32, 33, 34, 35, 36, 37, 38, 39],       [40, 41, 42, 43, 44, 45, 46, 47, 48, 49],       [50, 51, 52, 53, 54, 55, 56, 57, 58, 59],       [60, 61, 62, 63, 64, 65, 66, 67, 68, 69],       [70, 71, 72, 73, 74, 75, 76, 77, 78, 79],       [80, 81, 82, 83, 84, 85, 86, 87, 88, 89],       [90, 91, 92, 93, 94, 95, 96, 97, 98, 99]])

We can clip it with cropND(a, (5,5)), you will get:

>>> cropND(a, (5,5))array([[33, 34, 35, 36, 37],       [43, 44, 45, 46, 47],       [53, 54, 55, 56, 57],       [63, 64, 65, 66, 67],       [73, 74, 75, 76, 77]])

It not only works with 2D image but also 3D image.

Have a nice day.

Thanks, Divakar.

Your answer got me going the right direction. I came up with this using negative slice offsets to count 'from the end':

def cropimread(crop, xcrop, ycrop, fn):    "Function to crop center of an image file"    img_pre= msc.imread(fn)    if crop:        ysize, xsize, chan = img_pre.shape        xoff = (xsize - xcrop) // 2        yoff = (ysize - ycrop) // 2        img= img_pre[yoff:-yoff,xoff:-xoff]    else:        img= img_pre    return img