How to get image objects from Picture File Stream by cv2

I need to do some image processing recently. When I get the picture stream after a series of protocol parsing and decoding, the next step is to use cv2 to process the picture. First, I need to get the picture object of cv2.
generally
import cv2
img = cv2.imread ("test.jpg")
so you can get the img object directly.
but I want to convert directly in memory. I don"t want to write to the file and read it in cv2, so I can skip the step of image caching.
with the help of the great gods, I have been tinkering with it for a long time. How can it be realized?
such as
fc= open ("test.jpg", "rb"). Read ()
) when fc is the file stream of the image, which is the same as the file stream I processed.
so how do I read it in cv2 now?
cv2.imread (fc) is definitely wrong.

fc= open("test.jpg","rb").read() -sharpfcnumpy -sharpfcascii -sharpnpfcascii) -sharpfc100fileNPArry100 fileNPArray = np.fromstring(fc, np.uint8) -sharpcv2.IMREAD_COLOR -sharpcv2.IMREAD_GRAYSCALE img = cv2.imdecode(fileNPArray ,cv2.IMREAD_COLOR) -sharpimg cv2.imread()

>>> import cv -sharp cv2 >>> url='https://img.codeshelper.com/upload/img/2022/03/16/hl222fknxhx4070.jpg' >>> def get_net_img(url): img = None cap=cv.VideoCapture(url) if cap.isOpened(): ret,img=cap.read() return img >>> img = get_net_img(url) >>> cv.imshow('photo', img) >>>

-sharp 2 import cv import requests as req import numpy as np url2 = r'https://avatar-static.segmentfault.com/253/210/2532103026-593f867b9c17e' rsp = req.get(url2) cont = rsp.content buf = np.asarray(bytearray(cont), dtype="uint8") -sharp numpy img = cv.imdecode(buf, cv.IMREAD_COLOR) -sharp cv.imshow('pic', img) -sharp