You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
167 lines
6.8 KiB
167 lines
6.8 KiB
# 导入代码依赖
|
|
import time
|
|
import av
|
|
import os
|
|
import cv2
|
|
import torch
|
|
import numpy as np
|
|
import threading
|
|
import importlib.util
|
|
from core.DBManager import mDBM,DBManager
|
|
from myutils.MyLogger_logger import LogHandler
|
|
from myutils.ConfigManager import myCongif
|
|
|
|
|
|
|
|
class VideoCaptureWithFPS:
|
|
def __init__(self, src=0, target_fps=10):
|
|
self.cap = cv2.VideoCapture(src)
|
|
self.target_fps = target_fps
|
|
self.frame_interval = 1.0 / target_fps
|
|
self.last_frame_time = time.time()
|
|
|
|
def read(self):
|
|
current_time = time.time()
|
|
elapsed_time = current_time - self.last_frame_time
|
|
|
|
if elapsed_time < self.frame_interval: #小于间隔时间会休眠
|
|
time.sleep(self.frame_interval - elapsed_time)
|
|
|
|
self.last_frame_time = time.time()
|
|
ret, frame = self.cap.read()
|
|
return ret, frame
|
|
|
|
def release(self):
|
|
self.cap.release()
|
|
self.cap = None
|
|
|
|
|
|
class ModelManager:
|
|
def __init__(self):
|
|
self.verify_list = {}
|
|
self.bRun = True
|
|
self.logger = LogHandler().get_logger("ModelManager")
|
|
# 本地YOLOv5仓库路径
|
|
self.yolov5_path = myCongif.get_data("yolov5_path")
|
|
self.buflen = myCongif.get_data("buffer_len")
|
|
|
|
def _open_view(self,url,itype): #打开摄像头 0--USB摄像头，1-RTSP,2-海康SDK
|
|
if itype == 0:
|
|
cap = VideoCaptureWithFPS(int(url))
|
|
elif itype == 1:
|
|
cap = VideoCaptureWithFPS(url)
|
|
else:
|
|
raise Exception("视频参数错误！")
|
|
return cap
|
|
|
|
def _import_model(self,model_name,model_path):
|
|
'''根据路径，动态导入模块'''
|
|
if os.path.exists(model_path):
|
|
module_spec = importlib.util.spec_from_file_location(model_name, model_path)
|
|
module = importlib.util.module_from_spec(module_spec)
|
|
module_spec.loader.exec_module(module)
|
|
else:
|
|
self.logger.error("{}文件不存在".format(model_path))
|
|
return None
|
|
return module
|
|
|
|
def dowork_thread(self,channel_id):
|
|
'''一个通道一个线程，关联的模型在一个线程检测'''
|
|
cap = None
|
|
#查询关联的模型 --- 在循环运行前把基础数据都准备好
|
|
myDBM = DBManager()
|
|
myDBM.connection()
|
|
strsql = (f"select t1.model_id,t1.check_area,t1.polygon ,t2.duration_time,t2.proportion,t2.model_path "
|
|
f"from channel2model t1 left join model t2 on t1.model_id = t2.ID where t1.channel_id ={channel_id};")
|
|
myModels = myDBM.do_select(strsql)
|
|
#加载模型 --- 是不是要做个限制，一个视频通道关联算法模块的上限 --- 关联多了一个线程执行耗时较多，造成帧率太低，或者再多线程并发 #？
|
|
myModle_list = [] #存放模型对象List
|
|
myModle_data = [] #存放检测参数
|
|
for model in myModels:
|
|
#基于基类实例化模块类
|
|
m = self._import_model("",model[5]) #动态加载模型 -- 待完善
|
|
myModle_list.append(m)
|
|
myModle_data.append(model)
|
|
|
|
#开始循环检测
|
|
#print(mydata[0],mydata[1],mydata[2],mydata[3]) # url type tag img_buffer
|
|
#[url,type,True,img_buffer]
|
|
while self.verify_list[channel_id][2]: #基于tag 作为运行标识。 线程里只是读，住线程更新，最多晚一轮，应该不用线程锁。需验证
|
|
if not cap: #还没连接视频源
|
|
try:
|
|
cap = self._open_view(self.verify_list[channel_id][0],self.verify_list[channel_id][1])
|
|
except:
|
|
self.logger.error("参数错误，终止线程")
|
|
return
|
|
ret,frame = cap.read()
|
|
if not ret:
|
|
self.logger.warning("view disconnected. Reconnecting...")
|
|
cap.release()
|
|
cap = None
|
|
time.sleep(myCongif.get_data("cap_sleep_time"))
|
|
continue #没读到画面继续
|
|
#图像处理
|
|
img = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
|
|
# img = frame.to_ndarray(format="bgr24")
|
|
#检查布防时间是否在布防时间内
|
|
# 使用 模型 进行目标检测
|
|
i_warn_count = 0
|
|
for i in range(len(myModle_list)):#如果比较多可以考虑做线程进行检测
|
|
model = myModle_list[i]
|
|
data = myModle_data[i]
|
|
# 进行模型检测
|
|
detections,bwarn,warntext = model.verify(img,data)
|
|
#对识别结果要部要进行处理
|
|
if bwarn: # 整个识别有产生报警
|
|
# 绘制报警文本
|
|
cv2.putText(img, 'Intruder detected!', (50, (i_warn_count+1)*50),
|
|
cv2.FONT_HERSHEY_SIMPLEX, 1,(0, 0, 255), 2)
|
|
i_warn_count += 1
|
|
# 保存报警信息？ --- 待完成
|
|
self.save_warn()
|
|
# 推送报警？ --- 待完成
|
|
self.send_warn()
|
|
# 将检测结果图像转换为帧 -- 需要确认前面对img的处理都是累加的。
|
|
new_frame = av.VideoFrame.from_ndarray(img, format="rgb24")
|
|
# 分析图片放入内存中 ---
|
|
#new_frame = cv2.resize(new_frame, (640, 480)) # 降低视频分辨率
|
|
self.verify_list[channel_id][3].append(new_frame)
|
|
if len(self.verify_list[channel_id]) > self.buflen: # 保持缓冲区大小不超过10帧
|
|
self.verify_list[channel_id][3].pop(0)
|
|
self.logger.debug("drop one frame!")
|
|
|
|
|
|
|
|
def start_work(self,channel_id=0):
|
|
'''算法模型是在后台根据画面实时分析的'''
|
|
if channel_id ==0:
|
|
strsql = "select id,ulr,type from channel where is_work = 1;" #要考虑布防和撤防开关的的调整
|
|
else:
|
|
strsql = f"select id,ulr,type from channel where is_work = 1 and id = {channel_id};" #单通道启动检测线程
|
|
datas = mDBM.do_select(strsql)
|
|
for data in datas:
|
|
img_buffer = []
|
|
run_data = [data[1],data[2],True,img_buffer]
|
|
self.verify_list[data[0]] = run_data #需要验证重复情况
|
|
th_chn = threading.Thread(target=self.dowork_thread, args=(data[0],)) #一个视频通道一个线程，线程句柄暂时部保留
|
|
th_chn.start()
|
|
|
|
|
|
def stop_work(self,channel_id=0):
|
|
if channel_id ==0: #所有线程停止
|
|
for data in self.verify_list:
|
|
data[2] = False
|
|
del data[3]
|
|
time.sleep(2)
|
|
self.verify_list.clear()
|
|
else:
|
|
data = self.verify_list[channel_id]
|
|
data[2] = False
|
|
del data[3]
|
|
time.sleep(1)
|
|
del self.verify_list[channel_id]
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
ModelManager().start_work()
|