CODEWITHSUNDEEP
pythonimagenumpyopencv
Tawatchai WONGSANSUK
Tawatchai WONGSANSUK

Reputation: 21

How to convert image to bytearray?

I want create a video steaming on socket, but I can't convert an image to bytearray. why i can't convert image form camera to bytearray?

CHUNK=1024 lnF = 640*480*3    
wvs = WebcamVideoStream(0).start()    
while True: for x in range(1):
        try:
            frame = wvs.read()
            cv2_im = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            # cv2.imshow('frame',cv2_im)
            # if cv2.waitKey(1) & 0xFF == ord('q'):
            #     break
            frame = cv2.resize(frame, (640, 480))
            # print("frame :", frame)
            frame = np.array(frame, dtype = np.uint8).reshape(1, lnF)
            # print("frame :",frame)
            jpg = bytearray(frame)
            # print("jpg_as_text", jpg_as_text)
            # print("ACCEP")
        except Exception as e:
            print(e)

Upvotes: 2

Views: 1288

Answers (2)

Wouterr
Wouterr

Reputation: 516

An cv2 image is stored as an numpy.ndarray. To get the byte representation of an numpy.ndarray you can just use the numpy.ndarray.tobytes method to convert you image. In your code it would look something like this:

jpg = frame.tobytes()

The length of the bytes when using the .tobytes method is actually a bit less when using is actually a bit smaller than the pickle.dumps method and a lot faster.

Upvotes: 1

Gulzar
Gulzar

Reputation: 28074

Try this:

 import pickle

 frame = wvs.read()
 cv2_im = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
 frame = cv2.resize(frame, (640, 480))
 frame = np.array(frame, dtype = np.uint8).reshape(1, lnF)

 # packing:

 pickled = pickle.dumps(frame)
 byte_array_pickle = str.encode(pickled)

and

#unpacking

unicode_pickle = incoming_data.decode()
original_image = pickle.loads(unicode_pickle)

Upvotes: 0

Related Questions