Change CPU to GPU with OpenCV Python

Python
, ,
1

I use Jetson Nano with OpenCV. When I run the Mobilenet SSD object detection program at 1 FPS, when I check the GPU and CPU monitoring I see that it only uses the CPU.

The CPU goes from 0 to more than 50%. The GPU goes from 0 to 6%. I think that’s the problem.

My code:

import cv2
from datetime import datetime
import numpy as np
from centroidtracker.centroidtracker import CentroidTracker
import imutils

# Lista de todos los ID que concuerdan con el dueño (por si cambia el objectID)
owner_ids = []
objectId_list = []

protopath = "models/MobileNetSSD_deploy.prototxt"
modelpath = "models/MobileNetSSD_deploy.caffemodel"
detector = cv2.dnn.readNetFromCaffe(prototxt=protopath, caffeModel=modelpath)

CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat",
           "bottle", "bus", "car", "cat", "chair", "cow", "diningtable",
           "dog", "horse", "motorbike", "person", "pottedplant", "sheep",
           "sofa", "train", "tvmonitor"]
 
tracker = CentroidTracker(maxDisappeared=80, maxDistance=90)

def non_max_suppression_fast(boxes, overlapThresh):
    try:
        if len(boxes) == 0:
            return []

        if boxes.dtype.kind == "i":
            boxes = boxes.astype("float")

        pick = []

        x1 = boxes[:, 0]
        y1 = boxes[:, 1]
        x2 = boxes[:, 2]
        y2 = boxes[:, 3]

        area = (x2 - x1 + 1) * (y2 - y1 + 1)
        idxs = np.argsort(y2)

        while len(idxs) > 0:
            last = len(idxs) - 1
            i = idxs[last]
            pick.append(i)

            xx1 = np.maximum(x1[i], x1[idxs[:last]])
            yy1 = np.maximum(y1[i], y1[idxs[:last]])
            xx2 = np.minimum(x2[i], x2[idxs[:last]])
            yy2 = np.minimum(y2[i], y2[idxs[:last]])

            w = np.maximum(0, xx2 - xx1 + 1)
            h = np.maximum(0, yy2 - yy1 + 1)

            overlap = (w * h) / area[idxs[:last]]

            idxs = np.delete(idxs, np.concatenate(([last],
                                                   np.where(overlap > overlapThresh)[0])))

        return boxes[pick].astype("int")
    except Exception as e:
        print("Exception occurred in non_max_suppression : {}".format(e))

def main():
    #cap = cv2.VideoCapture("http://127.0.0.1:5000/video_feed")
    cap = cv2.VideoCapture(0)

    fps_start_time = datetime.now()
    fps = 0
    total_frames = 0

    while True:
        # Grab a single frame of video
        ret, frame = cap.read()
        #frame = imutils.resize(frame, width=400)
        frame = cv2.resize(frame, (0, 0), fx=0.25, fy=0.25)
    
        stop = "false"
        if stop == "false":
            total_frames = total_frames + 1
            (H, W) = frame.shape[:2]
            blob = cv2.dnn.blobFromImage(frame, 0.007843, (W, H), 127.5)
            detector.setInput(blob)
            person_detections = detector.forward()
            rects = []
            for i in np.arange(0, person_detections.shape[2]):
                confidence = person_detections[0, 0, i, 2]
                if confidence > 0.6: #PRECISIÓN DE DETECCIÓN DE OBJETOS
                    idx = int(person_detections[0, 0, i, 1])

                    if CLASSES[idx] != "person":
                        continue

                    person_box = person_detections[0, 0, i, 3:7] * np.array([W, H, W, H])
                    (startX, startY, endX, endY) = person_box.astype("int")
                    rects.append(person_box)

            boundingboxes = np.array(rects)
            boundingboxes = boundingboxes.astype(int)
            rects = non_max_suppression_fast(boundingboxes, 0.3)

            objects = tracker.update(rects)
            for (objectId, bbox) in objects.items():
                x1, y1, x2, y2 = bbox
                x1 = int(x1)
                y1 = int(y1)
                x2 = int(x2)
                y2 = int(y2)

                # Draw a point for the center of the object
                center_object = ((x1 + x2) // 2, (y1 + y2) // 2)
                
                # Draw the bounding box
                cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 0, 255), 2)
                #cv2.circle(frame, center_object, 2, (0, 0, 255), 2)

                #Update database
                update_data(center_object, "center", "vision_body_position")
                update_data(x1, "x1", "vision_body_position")
                update_data(y1, "y1", "vision_body_position")
                update_data(x2, "x2", "vision_body_position")
                update_data(y2, "y2", "vision_body_position")
                update_data(objectId, "objectId", "vision_body_position")

                text = f"ID: {objectId}, C:{center_object}"
                cv2.putText(frame, text, (x1, y1-5), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 0, 255), 2)           
               
            fps_end_time = datetime.now()
            time_diff = fps_end_time - fps_start_time
            if time_diff.seconds == 0:
                fps = 0.0
            else:
                fps = (total_frames / time_diff.seconds)

            fps_text = "FPS: {:.2f}".format(fps)

            cv2.putText(frame, fps_text, (5, 30), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 0, 255), 1)

        
        cv2.imshow("Vision", frame)
        key = cv2.waitKey(1)
        if key == ord('q'):
            break

    cv2.destroyAllWindows()

if __name__ == '__main__':
    main()

I am new, can you give me some recommendations to change it to GPU?