Convert grayscale YUV to an image (2D matrix)

Question

I'm trying to convert a grayscale yuv frame (from a video) to an image.

So I extracted the y component, which is a 2D matrix and saved it, but the result was not the image. I guess there should be some kind of transformation in order to get the right image?

Here's my code to extract frames from the video:

def yuv_get_frames_list(grayscale=False, frames_number=50):
  width = 1920  # [pixels] - Frame Width
  height = 1080  # [pixels] - Frame Height
  start_frame = 250
  end_frame = start_frame + frames_number
  bits_per_color = 8
  chroma_sampling = "420"

  input_file_name = "externals/Bosphorus_1920x1080_120fps_420_8bit_YUV.yuv"

  frames_lst = []
  for frame_idx in range(start_frame, end_frame):
    frame = yuv_read_frame(input_file_name, width, height, frame_idx, bits_per_color, chroma_sampling, grayscale)
    frames_lst.append(frame)

return frames_lst

So then, I take a frame (the first one), extract frame.y, transpose and save it with Pillow, but here's the result (it should be an image of a wolf):

Note that y.shape is (1920,1080), where cb,cr are of shape y.shape // 2 with values of 128 (for grayscale). Also note that when I write the frames to y4m, the video is presented correctly.

Thanks for your help :)