由於想針對影像做處理，使用 loop 去拿image 裡的pixel 使用 image.getpixel 似乎太沒有效率，於是使用 numpy.asarray() 把 PIL image object 轉成 array.
使用的範例圖片 width: 921, height 1001, 服用下面的 code 後：
import PIL
import numpy
from PIL import Image
im = Image.open('U_51234.bmp')
data = numpy.asarray(im)
print("shape:", data.shape)
print("100,150:",data[100,150])
print("100,150:",data[100][150])
取得 shape: (1001, 921)
所以是先 h 再 w, 如果要直接 access array 需要先放 y axis ，再使用 x axis, 與原先的 getpixel(x,y) 順序會相反。
Convert between PIL image and NumPy ndarray
image = Image.open(“max-demo.jpg”) # image is a PIL image
array = numpy.array(image) # array is a numpy array
image2 = Image.fromarray(array) # image2 is a PIL image
Convert between PIL image and PyOpenCV matrix
image = Image.open(“max-demo.jpg”) # image is a PIL image
mat = pyopencv.Mat.from_pil_image(image) # mat is a PyOpenCV matrix
image2 = mat.to_pil_image() # image2 is a PIL image
Convert between OpenCV image and NumPy ndarray
cimg = cv.LoadImage("max-demo.jpg", cv.CV_LOAD_IMAGE_COLOR) # cimg is a OpenCV image
pimg = Image.fromstring("RGB", cv.GetSize(cimg), cimg.tostring()) # pimg is a PIL image
array = numpy.array(pimg) # array is a numpy array
pimg2 = cv.fromarray(array) # pimg2 is a OpenCV image