AI Camera - IMX500 • Re: CCTV security environment

Code:

#!/usr/bin/env python3"""Based on Example module for Hailo Detection."""import argparseimport cv2from picamera2 import MappedArray, Picamera2, Previewfrom picamera2.devices import Hailofrom picamera2.encoders import H264Encoderfrom picamera2.outputs import CircularOutput2, PyavOutputfrom libcamera import controlsimport timeimport osimport datetime# detection objectsobjects = ["clock","person"]# set variablesv_width      = 1456  # video widthv_height     = 1088  # video heightv_length     = 5     # seconds, minimum video lengthpre_frames   = 5     # seconds, defines length of pre-detection bufferfps          = 25    # video frame rateshow_detects = 1     # show detections on video, 1 = yes, 0 = no# initialiseUsers  = []Users.append(os.getlogin())user   = Users[0]h_user = "/home/" + os.getlogin( )def extract_detections(hailo_output, w, h, class_names, threshold=0.5):    """Extract detections from the HailoRT-postprocess output."""    results = []    for class_id, detections in enumerate(hailo_output):        for detection in detections:            score = detection[4]            if score >= threshold:                y0, x0, y1, x1 = detection[:4]                bbox = (int(x0 * w), int(y0 * h), int(x1 * w), int(y1 * h))                results.append([class_names[class_id], bbox, score])    return resultsdef draw_objects(request):    global show_detects    current_detections = detections    if current_detections and show_detects == 1:        with MappedArray(request, "main") as m:            for class_name, bbox, score in current_detections:                x0, y0, x1, y1 = bbox                label = f"{class_name} %{int(score * 100)}"                cv2.rectangle(m.array, (x0, y0), (x1, y1), (0, 255, 0, 0), 4)                cv2.putText(m.array, label, (x0 + 5, y0 + 45),                            cv2.FONT_HERSHEY_SIMPLEX, 2, (0, 255, 0, 0), 3, cv2.LINE_AA)          if __name__ == "__main__":    # Parse command-line arguments.    parser = argparse.ArgumentParser(description="Detection Example")    parser.add_argument("-m", "--model", help="Path for the HEF model.",                        default="/usr/share/hailo-models/yolov8s_h8l.hef")    parser.add_argument("-l", "--labels", default="/home/" + user + "/picamera2/examples/hailo/coco.txt",                        help="Path to a text file containing labels.")    parser.add_argument("-s", "--score_thresh", type=float, default=0.5,                        help="Score threshold, must be a float between 0 and 1.")    args = parser.parse_args()    # Get the Hailo model, the input size it wants, and the size of our preview stream.    with Hailo(args.model) as hailo:        model_h, model_w, _ = hailo.get_input_shape()        video_w, video_h    = v_width,v_height        # Load class names from the labels file        with open(args.labels, 'r', encoding="utf-8") as f:            class_names = f.read().splitlines()        # The list of detected objects to draw.        detections = None        # Configure and start Picamera2.        main  = {'size': (video_w, video_h), 'format': 'YUV420'}        lores = {'size': (model_w, model_h), 'format': 'RGB888'}        controls2 = {'FrameRate': fps}        picam2 = Picamera2()        config = picam2.create_preview_configuration(main, lores=lores, controls=controls2)        picam2.configure(config)        encoder = H264Encoder(2000000)        circular = CircularOutput2(buffer_duration_ms=5000)        picam2.start_preview(Preview.QTGL, x=0, y=0, width=480, height=480)        picam2.start_recording(encoder, circular)        encoding = False        if show_detects == 1:            picam2.pre_callback = draw_objects                    # Process each low resolution camera frame.        while True:                # capture frame                frame = picam2.capture_array('lores')                # Run inference on the preprocessed frame                results = hailo.run(frame)                                # Extract detections from the inference results                detections = extract_detections(results, video_w, video_h, class_names, args.score_thresh)                # detection                for d in range(0,len(objects)):                    if len(detections) != 0:                        value = float(detections[0][2])                        obj = detections[0][0]                        if value > args.score_thresh and value < 1 and obj == objects[d]:                            startrec = time.monotonic()                            # start recording                            if not encoding:                                now = datetime.datetime.now()                                timestamp = now.strftime("%y%m%d_%H%M%S")                                circular.open_output(PyavOutput(h_user + "/Videos/" + timestamp +".mp4"))                                encoding = True                                print("New  Detection",timestamp,objects[d],timestamp + ".mp4")                                # save lores image                                cv2.imwrite(h_user + "/Pictures/" + str(timestamp) + ".jpg",frame)                # stop recording                if encoding and (time.monotonic() - startrec > v_length):                    now = datetime.datetime.now()                    timestamp2 = now.strftime("%y%m%d_%H%M%S")                    print("Stopped Record", timestamp2)                    circular.close_output()                    encoding = False