Next up: I want to create a waterfall showing the power distribution over the recorded bandwidth over time.

To do so, I save all the spectrums I calculate in one array. The it is as simple as putting that array in a file.

imagelist = []
    for arr in image:
        thislist = []
        for dat in arr:
            #dat = 10 * math.log10(dat)
            thislist.append(mymap(dat, min_pow, max_pow, 0, 255))
        imagelist.append(thislist)
    largearray = np.array(imagelist, np.ubyte)

    im = Image.fromarray(largearray, mode='L')
    im.save("ichbineinwasserfall.bmp")

My main problem was, that I didn't set the data type of the numpy array so the conversion with pillow into an image as a byte array failed.

This is the result for 94.7MHz. The spectrum is shown in a lot of other images I uploaded so I don't want to post it yet again.

By the way: I use 94.7 as center frequency mostly because there is a radio station near me so I have a good signal to practice with 24/7. 

The next step is to combine the waterfall and the graph in one window.