Rotating images in Python - Improving Quality - Scipy

Question

as the title says, I'm trying to rotate an image in Python, using the Scipy package. I receive as an input an image, with an specific size, for example, 512 x 512, and I have to perform some rotations over it and to crop a proportion of it. So, this is my code so far:

for i in range(0, 90, 10):
    image_base = Dt.Data(open_image=True, rotation=i)
    Dv.DataVisualization.plot_data(image_base, '../Visualizations/ProcessedImages/'+str(i))

def plot_data(data, file_path):
    """
    Plot the data
    :param data : the data to be plotted
    :param file_path : the name of the file
    """
    output = Image.fromarray(numpy.uint8(data.data * 255))
    output.save(file_path + '.png', 'PNG')

class Data(object):
    def __init__(self, open_image=False, rotation=None):
        """
        The Data constructor
        """
        self.data = misc.imread('../TestImages/board.jpg', flatten=True)
        self.data /= 255.0
        x, y = self.data.shape

        if rotation != 0:
            self.data = ndimage.rotate(self.data, rotation, reshape=False)
            self.data = self.data[x / 4: - x / 4, y / 4: - y / 4]

This is my original image (256x256):

And these are some outputs there I got (128x128):

As you can see, the edges are pretty bad, and this is not acceptable for what I have to do. I'm wondering why this happens and if there is a way to get rid of it.

Thank you in advance.

Answer 1

Ok, I was able to solve this using the parameter:

order : int, optional

The order of the spline interpolation, default is 3. The order has to be in the range 0-5.

Thank you for your time. (: