Inference code for running Tensorflow lite model on google coral edge tpu using pi camera

inference.py

from edgetpu.detection.engine import DetectionEngine
from PIL import Image
from picamera.array import PiRGBArray
from picamera import PiCamera
import argparse
import time
import cv2
# construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-m", "--model", required=True,
                help="path to TensorFlow Lite object detection model")
ap.add_argument("-c", "--confidence", type=float, default=0.3,
                help="minimum probability to filter weak detections")
args = vars(ap.parse_args())

print("[INFO] parsing class labels...")
camera = PiCamera()
camera.resolution = (320, 240)
camera.framerate = 30
rawCapture = PiRGBArray(camera, size=(320, 240))

# time.sleep(1)

display_window = cv2.namedWindow("detection")


print("[INFO] loading Coral model...")
model = DetectionEngine(args["model"])
count = 0
start = time.time()

detect = True

for frame in camera.capture_continuous(rawCapture, format="bgr", use_video_port=True):
    # grab the frame from the threaded video stream and resize it
    # to have a maximum width of 500 pixels
    image = cv2.cvtColor(frame.array, cv2.COLOR_BGR2RGB)
    image = Image.fromarray(image)
    orig = frame.array

    # prepare the frame for object detection by converting (1) it
    # from BGR to RGB channel ordering and then (2) from a NumPy
    # array to PIL image format
    # make predictions on the input frame
    if detect:
        results = model.detect_with_image(
            image, threshold=args["confidence"], keep_aspect_ratio=True, relative_coord=False)
        for r in results:
            # extract the bounding box and box and predicted class label
            box = r.bounding_box.flatten().astype("int")
            (startX, startY, endX, endY) = box
            label = r.label_id
            # draw the bounding box and label on the image
            cv2.rectangle(orig, (startX, startY), (endX, endY),
                          (0, 255, 0), 2)
            cv2.putText(orig, str(label), (startX, startY - 15),
                        cv2.FONT_HERSHEY_PLAIN, 1, (0, 255, 0), 2)
            y = startY - 15 if startY - 15 > 15 else startY + 15
    # show the output frame and wait for a key press
    cv2.imshow("detection", orig)
    key = cv2.waitKey(1) & 0xFF
    count += 1
    if (time.time() - start) > 1:
        print("FPS: ", count / (time.time() - start))
        count = 0
        start = time.time()
    # print(count / (time.time() - start))
    rawCapture.truncate(0)
    # if the `q` key was pressed, break from the loop
    if key == ord("q"):
        break
    if key == ord("c"):
        detect = not detect
# do a bit of cleanup
cv2.destroyAllWindows()
camera.close()