import gradio as gr import cv2 import tempfile from ultralytics import YOLO import io import base64 from utils import get_coordinates_from_exif, covert_image_tobase64,get_adress def yolov10_inference(collection,image, video, model_id, image_size, conf_threshold): model = YOLO(f'runs/{model_id}.pt') lat, lon,date_create = get_coordinates_from_exif(image) # if lat is None or lon is None: # gr.Warning('Không tìm thấy tọa độ GPS trong hình ảnh') # return "Không tìm thấy tọa độ GPS trong hình ảnh", None if image: results = model.predict(source=image, imgsz=(640,640), conf=conf_threshold,agnostic_nms=True, show_conf=False) bboxes = results[0].boxes.xywh.cpu().numpy().tolist() clss = results[0].boxes.cls.cpu().numpy().tolist() annotated_image = results[0].plot() img_str, im_rgb = covert_image_tobase64(annotated_image) # Lưu trữ hình ảnh và bbox vào MongoDB if len(bboxes) > 0 and lat is not None and lon is not None and model_id == 'Yolov8n_4': address = get_adress(lat, lon) collection.update_one({"image": img_str}, { "$set": {"image": img_str, "latitude": lat, "longitude": lon, "bboxes": bboxes, "class":clss, "address":address,"date_create":date_create}}, True) # return annotated_image[:, :, ::-1], None return im_rgb, None else: # video_path = tempfile.mktemp(suffix=".webm") # with open(video_path, "wb") as f: # with open(video, "rb") as g: # f.write(g.read()) cap = cv2.VideoCapture(video) fps = cap.get(cv2.CAP_PROP_FPS) frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) output_video_path = tempfile.mktemp(suffix=".webm") out = cv2.VideoWriter(output_video_path, cv2.VideoWriter_fourcc(*'vp80'), fps, (frame_width, frame_height)) while cap.isOpened(): ret, frame = cap.read() if not ret: break results = model.track(frame, persist=True, conf=0.5, imgsz=(640,640)) # Check if tracking IDs are available if hasattr(results[0].boxes, 'id') and results[0].boxes.id is not None: annotated_frame = results[0].plot() boxes = results[0].boxes.xyxy.cpu().numpy().astype(int) ids = results[0].boxes.id.cpu().numpy().astype(int) conf = max(results[0].boxes.conf.cpu().numpy()) lat, lon,date_create = get_coordinates_from_exif(frame) out.write(annotated_frame) # re None, output_video_path,annotated_frame # results[0].save(filename=f'img_result/VID_20240513_171727_{ids[0]}_{conf}.jpg') # # Draw boxes and IDs on the frame # for box, id in zip(boxes, ids): # cv2.rectangle(frame, (box[0], box[1]), (box[2], box[3]), (0, 255, 0), 2) # cv2.putText(frame, f"Id {id}", (box[0], box[1]), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2) else: out.write(frame) print("Tracking IDs not available.") cap.release() out.release() return None, output_video_path