openCV9_ROI_bounceBox.py

#program launches the rpi cam with jetson nano
#playing around with Regions of Interest, matrix and data conversion.
#Programtakes a frame, looks at a roi, converts it to grayscale, then coverts the data
#type back to colour, then inserts into original image

import cv2

print(cv2.__version__)

#Setting up the camera
dispW=640
dispH=480
flip=0
camSet='nvarguscamerasrc !  video/x-raw(memory:NVMM), width=3264, height=2464, format=NV12, framerate=21/1 ! nvvidconv flip-method='+str(flip)+' ! video/x-raw, width='+str(dispW)+', height='+str(dispH)+', format=BGRx ! videoconvert ! video/x-raw, format=BGR ! appsink'

cam = cv2.VideoCapture(camSet)

#Read camera until q key is pressed
while True:
    ret, frame = cam.read()
    roi = frame[50:250, 200:400].copy()
    roiGray = cv2.cvtColor(roi, cv2.COLOR_BGR2GRAY)
    roiGray = cv2.cvtColor(roiGray, cv2.COLOR_GRAY2BGR)
    frame[50:250,200:400] = roiGray
    cv2.imshow('roi', roi)
    cv2.imshow('piCam',frame)
    cv2.imshow('Gray', roiGray)
    cv2.moveWindow('piCam', 0, 0)
    cv2.moveWindow('roi', 705, 0)
    cv2.moveWindow('Gray', 705, 250)
    if cv2.waitKey(1) == ord('q'):
        break

#After q is pressed, release the camera back to the system
cam.release()
cv2.destroyAllWindows()