python
复制代码
# 实时检测跟踪模块，并将检测跟踪结果保存到数据库中
# -*- coding: utf-8 -*-
import argparse
import os
os.environ["OMP_NUM_THREADS"] = "1"
os.environ["OPENBLAS_NUM_THREADS"] = "1"
os.environ["MKL_NUM_THREADS"] = "1"
os.environ["VECLIB_MAXIMUM_THREADS"] = "1"
os.environ["NUMEXPR_NUM_THREADS"] = "1"
import sys
import platform
import platform
import numpy as np
from pathlib import Path
import torch
import torch.backends.cudnn as cudnn
from shapely.geometry import Polygon
import numpy as np
import matplotlib.pyplot as plt
import pymysql
import time
from datetime import datetime
import json
FILE = Path(__file__).resolve()
ROOT = FILE.parents[0] # yolov5 strongsort root directory
WEIGHTS = ROOT / 'weights'
if str(ROOT) not in sys.path:
sys.path.append(str(ROOT)) # add ROOT to PATH
if str(ROOT / 'yolov5') not in sys.path:
sys.path.append(str(ROOT / 'yolov5')) # add yolov5 ROOT to PATH
if str(ROOT / 'trackers' / 'strongsort') not in sys.path:
sys.path.append(str(ROOT / 'trackers' / 'strongsort')) # add strong_sort ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
from yolov5.models.common import DetectMultiBackend
from yolov5.utils.dataloaders import VID_FORMATS, LoadImages, LoadStreams
from yolov5.utils.general import (LOGGER, Profile, check_img_size, non_max_suppression, scale_boxes, check_requirements, cv2,
check_imshow, xyxy2xywh, increment_path, strip_optimizer, colorstr, print_args, check_file)
from yolov5.utils.torch_utils import select_device, time_sync
from yolov5.utils.plots import Annotator, colors, save_one_box
from trackers.multi_tracker_zoo import create_tracker
def get_connection():
"""创建并返回一个新的数据库连接。"""
# 数据库连接信息
host = 'localhost'
user = 'root'
password = '123456'
database = 'video_streaming_database'
return pymysql.connect(host=host, user=user, password=password, database=database)
def ensure_connection(connection):
"""确保连接有效。如果连接无效，则重新建立连接。"""
if connection is None or not connection.open:
print("Connection is invalid or closed. Reconnecting...")
return get_connection()
return connection
@torch.no_grad()
def run(
source='0',
yolo_weights=WEIGHTS / 'yolov5m.pt', # model.pt path(s),
reid_weights=WEIGHTS / 'osnet_x0_25_msmt17.pt', # model.pt path,
tracking_method='strongsort',
tracking_config=None,
imgsz=(640, 640), # inference size (height, width)
cam_ip = '192.168.31.97',
conf_thres=0.25, # confidence threshold
iou_thres=0.45, # NMS IOU threshold
max_det=1000, # maximum detections per image
device='0', # cuda device, i.e. 0 or 0,1,2,3 or cpu
show_vid=False, # show results
save_txt=False, # save results to *.txt
save_conf=False, # save confidences in --save-txt labels
save_crop=False, # save cropped prediction boxes
save_trajectories=False, # save trajectories for each track
save_vid=True, # save confidences in --save-txt labels
nosave=False, # do not save images/videos
classes=None, # filter by class: --class 0, or --class 0 2 3
agnostic_nms=False, # class-agnostic NMS
augment=False, # augmented inference
visualize=False, # visualize features
update=False, # update all models
project=ROOT / 'runs' / 'track', # save results to project/name
name='exp', # save results to project/name
exist_ok=False, # existing project/name ok, do not increment
line_thickness=2, # bounding box thickness (pixels)
hide_labels=False, # hide labels
hide_conf=False, # hide confidences
hide_class=False, # hide IDs
half=False, # use FP16 half-precision inference
dnn=False, # use OpenCV DNN for ONNX inference
vid_stride=1, # video frame-rate stride
retina_masks=False,
):
source = str(source)
is_file = Path(source).suffix[1:] in (VID_FORMATS)
is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://'))
webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file)
if is_url and is_file:
source = check_file(source)
if not isinstance(yolo_weights, list): # single yolo model
exp_name = yolo_weights.stem
elif type(yolo_weights) is list and len(yolo_weights) == 1: # single models after --yolo_weights
exp_name = Path(yolo_weights[0]).stem
else: # multiple models after --yolo_weights
exp_name = 'ensemble'
# 结果保存路径
project = os.path.join(os.path.dirname(source), (source.split("\\")[-1][:-4])) + "_det"
save_dir = increment_path(Path(project), exist_ok=exist_ok) # increment run
(save_dir / 'tracks' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir
# 载入模型
device = select_device(device)
model = DetectMultiBackend(yolo_weights, device=device, dnn=dnn, data=None, fp16=half)
stride, names, pt = model.stride, model.names, model.pt
imgsz = check_img_size(imgsz, s=stride) # check image size
if webcam:
show_vid = check_imshow()
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride)
nr_sources = len(dataset)
else:
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
nr_sources = 1
tracker_list = []
for i in range(nr_sources):
tracker = create_tracker(tracking_method, tracking_config, reid_weights, device, half)
tracker_list.append(tracker, )
if hasattr(tracker_list[i], 'model'):
if hasattr(tracker_list[i].model, 'warmup'):
tracker_list[i].model.warmup()
outputs = [None] * nr_sources
# Run tracking
seen, windows, dt = 0, [], (Profile(), Profile(), Profile(), Profile())
curr_frames, prev_frames = [None] * nr_sources, [None] * nr_sources
# 数据库连接信息
host = 'localhost'
user = 'root'
password = '123456'
database = 'video_streaming_database'
connection = pymysql.connect(host=host, user=user, password=password, database=database)
data = []
for frame_idx, (path, im, im0s, vid_cap, s) in enumerate(dataset):
start_time = time.time()
im_Original = im0s
# 隔帧操作，实际测试对跟踪计数影响很大
if frame_idx % 2 != 0:
im_Original_resieze = cv2.resize(im_Original, dsize=None, fx=0.5, fy=0.5, interpolation=cv2.INTER_CUBIC)
cv2.imwrite(os.path.join(str(save_dir), cam_ip + "_" + str(frame_idx + 1).zfill(8) + ".jpg"), im_Original_resieze)
continue
with dt[0]:
im = torch.from_numpy(im).to(device)
im = im.half() if half else im.float() # uint8 to fp16/32
im /= 255.0 # 0 - 255 to 0.0 - 1.0
if len(im.shape) == 3:
im = im[None] # expand for batch dim
with dt[1]:
pred = model(im, augment=augment, visualize=visualize)
with dt[2]:
pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
# 处理检测结果
for i, det in enumerate(pred): # detections per image
seen += 1
if webcam: # nr_sources >= 1
p, im0, _ = path[i], im0s[i].copy(), dataset.count
p = Path(p) # to Path
s += f'{i}: '
txt_file_name = p.name
save_path = str(save_dir / p.name) # im.jpg, vid.mp4, ...
else:
p, im0, _ = path, im0s.copy(), getattr(dataset, 'frame', 0)
p = Path(p) # to Pathf
# video file
if source.endswith(VID_FORMATS):
txt_file_name = p.stem
save_path = str(save_dir / p.name) # im.jpg, vid.mp4, ...
# folder with imgs
else:
txt_file_name = p.parent.name # get folder name containing current img
save_path = str(save_dir / p.parent.name) # im.jpg, vid.mp4, ...
curr_frames[i] = im0
s += '%gx%g ' % im.shape[2:] # print string
annotator = Annotator(im0, line_width=line_thickness, example=str(names))
if hasattr(tracker_list[i], 'tracker') and hasattr(tracker_list[i].tracker, 'camera_update'):
if prev_frames[i] is not None and curr_frames[i] is not None: # camera motion compensation
tracker_list[i].tracker.camera_update(prev_frames[i], curr_frames[i])
if det is not None and len(det):
det[:, :4] = scale_boxes(im.shape[2:], det[:, :4], im0.shape).round() # rescale boxes to im0 size
for c in det[:, 5].unique():
n = (det[:, 5] == c).sum() # detections per class
s += f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
with dt[3]:
outputs[i] = tracker_list[i].update(det.cpu(), im0)
# 处理跟踪结果
if len(outputs[i]) > 0:
for j, (output) in enumerate(outputs[i]):
bbox = output[0:4]
id = output[4]
cls = output[5]
conf = output[6]
bbox_x = int((output[0] + output[2]) / 2)
bbox_y = int((output[1] + output[3]) / 2)
bbox_w = int(output[2] - output[0])
bbox_h = int(output[3] - output[1])
if save_vid or save_crop or show_vid: # Add bbox to image
c = int(cls) # integer class
id = int(id) # integer id
label = None if hide_labels else (f'{id} {names[c]}' if hide_conf else \
(f'{id} {conf:.2f}' if hide_class else f'{id} {names[c]} {conf:.2f}'))
color = colors(c, True)
annotator.box_label(bbox, label, color=color)
if save_trajectories and tracking_method == 'strongsort':
q = output[7]
tracker_list[i].trajectory(im0, q, color=color)
if save_crop:
bbox = np.array(bbox)
if frame_idx % 12 == 0:
save_one_box(bbox.astype(np.int16), im_Original, file = save_dir / f'{id}' /
(cam_ip + "_"
+ str(frame_idx + 1).zfill(8) + "_"
+ str(id).zfill(4) + "_"
+ str(int(bbox_x)).zfill(4) + "_"
+ str(int(bbox_y)).zfill(4) + "_"
+ str(int(bbox_w)).zfill(4) + "_"
+ str(int(bbox_h)).zfill(4) + "_"
+ str(int(float(conf) * 10000))
+ f'.jpg'), BGR=True)
# 将检测跟踪中间结果保存到数据库中
connection = ensure_connection(connection) # 确保连接有效
# 获取当前日期和时刻
current_time = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
try:
with connection.cursor() as cursor:
# 插入数据的SQL语句
insert_sql = """
INSERT INTO new_detection_tracking_results_1 (camera_ip, frame_number, tracking_id, crop_image_path, event_datetime)
VALUES (%s, %s, %s, %s, %s);
"""
# 示例数据
data = [
(cam_ip,
int(frame_idx+1),
int(id),
save_dir / f'{id}' /
(cam_ip + "_"
+ str(frame_idx + 1).zfill(8) + "_"
+ str(id).zfill(4) + "_"
+ str(int(bbox_x)).zfill(4) + "_"
+ str(int(bbox_y)).zfill(4) + "_"
+ str(int(bbox_w)).zfill(4) + "_"
+ str(int(bbox_h)).zfill(4) + "_"
+ str(int(float(conf) * 10000))
+ f'.jpg'),
current_time)
]
# 执行插入操作
cursor.executemany(insert_sql, data)
connection.commit()
finally:
pass
else:
pass
# # 将检测跟踪的原图，标注图，检测结果保存到数据库中
im_Original_resieze = cv2.resize(im_Original, dsize=None, fx=0.5, fy=0.5, interpolation=cv2.INTER_CUBIC)
cv2.imwrite(os.path.join(str(save_dir), cam_ip + "_" + str(frame_idx + 1).zfill(8) + ".jpg"), im_Original_resieze)
# 保存检测跟踪结果到文件
if outputs[0] == None:
track_outputs = []
else:
track_outputs = [
[float(x[0] / 2), float(x[1] / 2), float(x[2] / 2), float(x[3] / 2), int(x[4]), float(x[6]), ""]
for x in outputs[0]
]
data_dict = {}
for row in track_outputs:
key = int(row[4])
value = row
data_dict[key] = value
json_output_path = os.path.join(str(save_dir), cam_ip + "_" + str(frame_idx + 1).zfill(8) + "_track.json")
with open(json_output_path, 'w') as json_file:
json.dump(data_dict, json_file, indent=4)
# 记录结束时间
end_time = time.time()
# 计算并打印运行时间
print(f"第{frame_idx}帧，程序运行时间: {end_time - start_time}秒")
if end_time - start_time >= 0.0833333333333333333333:
print(f"第{frame_idx}帧，程序运行时间: {end_time - start_time}秒")
if (end_time - start_time < 0.0833333333333333333333):
time.sleep(0.0833333333333333333333-end_time+start_time)
def parse_opt():
parser = argparse.ArgumentParser()
parser.add_argument('--yolo-weights', nargs='+', type=Path, default=R'/home/hitsz/yk_workspace/Yolov5_track/weights/train_citys_bdd_4S_crowdhuman_coco_labs_liucl_1215_no_freeze_no_freeze_yolov5m3/weights/v5m_861.pt', help='model.pt path(s)')
parser.add_argument('--reid-weights', type=Path, default=R'weights\osnet_x1_0_msmt17.pt')
parser.add_argument('--tracking-method', type=str, default='bytetrack', help='strongsort, ocsort, bytetrack')
parser.add_argument('--tracking-config', type=Path, default=None)
parser.add_argument('--source', type=str, default=R"02_output_0.mp4", help='file/dir/URL/glob, 0 for webcam')
# 下面为输入为摄像头视频流的参数设置
# parser.add_argument('--source', type=str, default=R'rtsp://admin:1234qwer@192.168.1.64:554/Streaming/Channels/101', help='file/dir/URL/glob, 0 for webcam')
parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w')
parser.add_argument('--cam_ip', type=str, default='192.168.31.97')
parser.add_argument('--conf-thres', type=float, default=0.45, help='confidence threshold')
parser.add_argument('--iou-thres', type=float, default=0.25, help='NMS IoU threshold')
parser.add_argument('--max-det', type=int, default=1000, help='maximum detections per image')
parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
parser.add_argument('--show-vid', default=False, action='store_true' , help='display tracking video results')
parser.add_argument('--save-txt', default=True, action='store_true', help='save results to *.txt')
parser.add_argument('--save-conf', default=True, action='store_true', help='save confidences in --save-txt labels')
parser.add_argument('--save-crop', default=True, action='store_true', help='save cropped prediction boxes')
parser.add_argument('--save-trajectories', default=True, action='store_true', help='save trajectories for each track')
parser.add_argument('--save-vid', default=True, action='store_true', help='save video tracking results')
parser.add_argument('--nosave', default=False, action='store_true', help='do not save images/videos')
parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --classes 0, or --classes 0 2 3')
parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS')
parser.add_argument('--augment', action='store_true', help='augmented inference')
parser.add_argument('--visualize', action='store_true', help='visualize features')
parser.add_argument('--update', action='store_true', help='update all models')
parser.add_argument('--project', default=R"/home/hitsz/yk_web/Yolov5_track/results/test_save_results1", help='save results to project/name')
parser.add_argument('--name', default='test', help='save results to project/name')
parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
parser.add_argument('--line-thickness', default=2, type=int, help='bounding box thickness (pixels)')
parser.add_argument('--hide-labels', default=False, action='store_true', help='hide labels')
parser.add_argument('--hide-conf', default=False, action='store_true', help='hide confidences')
parser.add_argument('--hide-class', default=False, action='store_true', help='hide IDs')
parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference')
parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference')
parser.add_argument('--vid-stride', type=int, default=1, help='video frame-rate stride')
parser.add_argument('--retina-masks', action='store_true', help='whether to plot masks in native resolution')
opt = parser.parse_args()
opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand
opt.tracking_config = ROOT / 'trackers' / opt.tracking_method / 'configs' / (opt.tracking_method + '.yaml')
print_args(vars(opt))
return opt
def main(opt):
check_requirements(requirements=ROOT / 'requirements.txt', exclude=('tensorboard', 'thop'))
run(**vars(opt))
if __name__ == "__main__":
opt = parse_opt()
main(opt)
import { createApp, createElementBlock } from 'vue';
import App from './App.vue';
import "@/assets/less/index.less";
import router from "@/router";
import ElementPlus from 'element-plus'
import 'element-plus/dist/index.css'
import * as ElementPlusIconsVue from '@element-plus/icons-vue'
import {createPinia} from 'pinia'
import "video.js/dist/video-js.css";
import "@/api/mock.js";
import api from '@/api/api'
import {useALLDataStore} from "@/stores"
function isRoute(to){
const routes = router.getRoutes();
// 检查是否有匹配的路由
return routes.some(route => {
// 处理动态路径匹配
const regex = new RegExp(`^${route.path.replace(/:\w+/g, '\\w+')}$`);
return regex.test(to.path);
});
}
const pinia = createPinia();
const app = createApp(App);
app.config.globalProperties.$api = api;
for (const [key, component] of Object.entries(ElementPlusIconsVue)) {
app.component(key, component)
}
app.use(pinia)
const store = useALLDataStore();
app.use(ElementPlus)
store.addMenu(router,"refresh")
app.use(router).mount("#app");