python - Convert PIL image to OpenCV2 image -

i'm using entirescreen=imagegrab.grab() create screengrab , conduct analyses opencv2 test whether screen contains template images. templates loaded template = cv2.imread(name,0).

i have following problem: when comparing screenshot templates need first save screenshot :

entirescreen.save('pics/screenshot.png', format='png')

and reload :

cv2.imread('screenshot.png',0)

otherwise following not work:

res = cv2.matchtemplate(img,template,method)

i error message one:

typeerror: image not numpy array, neither scalar

my question: how can convert screenshot entirescreen=imagegrab.grab() format compatible opencv2, without having save , reload cv2.imread.

on linux systems, 1 can use pyscreenshot docs state, replacement imagegrab module, works on windows only.

so, on linux system did -

import pyscreenshot imagegrab 

then, can grab screenshot , have access directly in memory space numpy array without having save onto disk , read imread, -

img = np.array(imagegrab.grab().convert('rgb')) 

this img used cv2.matchtemplate is.

---

a step-by-step sample run , verification of output -

in [32]: import pyscreenshot imagegrab  in [33]: img = np.array(imagegrab.grab().convert('rgb'))  in [34]: img.shape out[34]: (768, 1366, 3)  in [35]: img.dtype out[35]: dtype('uint8') 

here's sample run showing how cv2.matchtemplate used img -

in [41]: res = cv2.matchtemplate(img[:,:,0],img[10:40,50:80,0],cv2.tm_sqdiff)  in [42]: np.where(res<10) # 10 threshold matching out[42]: (array([10]), array([50]))