一天一元网站建设,手机网站自动适配,无人售货机,网站建设结论及体会1、搭建流媒体服务器
下载mediamtx 2、视频流直推 ffmpeg -stream_loop -1 -i DJI_20250109112715_0002_W.MP4 -r 30 -c:v libx264 -preset ultrafast -f flv rtmp://192.168.100.20:1935/live/test_chengdu1 3、硬件加速
如果硬件支持#xff0c;可以使用硬件加速编码器可以使用硬件加速编码器如 h264_nvenc、h264_vaapi 或 h264_qsv以减少 CPU 负载。 NVIDIA GPU-c:v h264_nvenc Intel GPU-c:v h264_qsv AMD GPU-c:v h264_amf 4、比特率控制 -b:v设置视频比特率。过高的比特率可能导致网络拥塞过低的比特率可能导致画质下降。根据网络带宽合理设置。 -maxrate 和 -bufsize限制最大比特率和缓冲区大小避免网络波动导致卡顿。 5、 网络缓冲
在FFmpeg命令中增加网络缓冲参数以应对网络不稳定。 -re 这个参数会让FFmpeg以原始速率读取输入而不是实时编码从而给网络传输留下更多的缓冲时间。 比如ffmpeg -i http://192.168.1.100:8080/video -c:v libx264 -s 1280x720 -b:v 2000k -r 15 -bufsize 4000k -crf 18 -hwaccel nvenc -re output.ffm 6、ffmpeg直推视频 ffmpeg -re -stream_loop -1 -i DJI_0180.MP4 -r 30 -c:v libx264 -f flv rtmp://192.168.100.20:1935/live/test_chengdu1 6、使用ffmpeg直接读取视频
frame_queue queue.Queue(maxsize1)
def read_video2(self, video_path, frame_queue):ffmpegReadCommand [ffmpeg,-i, video_path, # 输入流-f, image2pipe, # 输出格式为图像管道-pix_fmt, bgr24, # 像素格式为 BGR24-vcodec, rawvideo, # 视频编码为原始视频-]ffmpegReadProcess subprocess.Popen(ffmpegReadCommand, stdoutsubprocess.PIPE) width 1920 # 根据实际分辨率调整height 1080 # 根据实际分辨率调整try:while self.is_switchOn:t5 time.time()raw_frame ffmpegReadProcess.stdout.read(width * height * 3)if not raw_frame:print(无法读取帧)breakframe np.frombuffer(raw_frame, dtypeuint8).reshape((height, width, 3))t6 time.time()print(t6 - t5 , t6 - t5)try:frame_queue.put_nowait(frame)except queue.Full:# 如果队列满了丢弃旧帧以防止卡顿frame_queue.get_nowait()frame_queue.put_nowait(frame)except KeyboardInterrupt:print(视频流中断)# 清空队列可选如果只想处理最新帧if not frame_queue.empty():frame_queue.get(blockTrue, timeout50000)# 将最新帧放入队列frame_queue.put(frame)finally:# 释放资源ffmpegReadProcess.terminate()print(视频流结束)
7、最终代码
import cv2
import time
# import os
import numpy as np
import copy
import queue
import threading
from ultralytics import YOLO
import subprocess# 创建一个队列用于存储视频帧# os.environ[OPENCV_FFMPEG_READ_ATTEMPTS] 100000000000class multiDealImg(object):def __init__(self, model_path, rtmp):self.model YOLO(model_path)#self.frame_queue queue.Queue(maxsize1) # 设置队列的最大大小这里设置为1确保只处理最新的帧# self.show_queue queue.Queue(maxsize1)self.command [ffmpeg,-y,# -map, 0:v-f, rawvideo,-vcodec, rawvideo,-pix_fmt, bgr24,-s, {}x{}.format(1280,720), # 根据输入视频尺寸填写-r, 27,-i, -,# -c:v, libx264,-c:v, h264_nvenc,-b:v, 2M,-maxrate, 2M,-bufsize, 4M,-pix_fmt, yuv420p,# -preset, ultrafast,-f, flv,# -hwaccel, nvenc, -crf, 23,rtmp]self.pipe subprocess.Popen(self.command, stdinsubprocess.PIPE)self.is_switchOn Trueself.ffmpeg_command [ffmpeg,-y, # 覆盖输出文件-f,rawvideo,-vcodec,rawvideo,-pix_fmt,bgr24, # OpenCV读取的帧格式是BGR-s,{}x{}.format(1280, 720), # 根据输入视频尺寸填写-r,30,-i,-, # 输入来自管道stdin-c:v,libx264, # 使用H.264编码-pix_fmt,yuv420p, # 确保兼容性-crf,23, # 设置CRF值质量./out.mp4,]self.ffmpegSaveVideo subprocess.Popen(self.ffmpeg_command, stdinsubprocess.PIPE)self.save_queue queue.Queue(maxsize3)self.push_queue queue.Queue(maxsize3)# 视频读取函数def read_video2(self, video_path, frame_queue):ffmpegReadCommand [ffmpeg,-i, video_path, # 输入流-f, image2pipe, # 输出格式为图像管道-pix_fmt, bgr24, # 像素格式为 BGR24-vcodec, rawvideo, # 视频编码为原始视频-]ffmpegReadProcess subprocess.Popen(ffmpegReadCommand, stdoutsubprocess.PIPE) width 1920 # 根据实际分辨率调整height 1080 # 根据实际分辨率调整try:while self.is_switchOn:t5 time.time()raw_frame ffmpegReadProcess.stdout.read(width * height * 3)if not raw_frame:print(无法读取帧)breakframe np.frombuffer(raw_frame, dtypeuint8).reshape((height, width, 3))t6 time.time()print(t6 - t5 , t6 - t5)try:frame_queue.put_nowait(frame)except queue.Full:# 如果队列满了丢弃旧帧以防止卡顿frame_queue.get_nowait()frame_queue.put_nowait(frame)except KeyboardInterrupt:print(视频流中断)# 清空队列可选如果只想处理最新帧if not frame_queue.empty():frame_queue.get(blockTrue, timeout50000)# 将最新帧放入队列frame_queue.put(frame)finally:# 释放资源ffmpegReadProcess.terminate()print(视频流结束)def read_video(self, video_path, frame_queue):cap cv2.VideoCapture(video_path)total_frames int(cap.get(cv2.CAP_PROP_FRAME_COUNT))print(fTotal frames in the video: {total_frames})frame_counter 0fps cap.get(cv2.CAP_PROP_FPS)while cap.isOpened() and self.is_switchOn:#time.sleep(0.05)# print(111)t5 time.time()ret, frame cap.read()frame_counter 1#循环读取视频条件判断if frame_counter int(cap.get(cv2.CAP_PROP_FRAME_COUNT)):frame_counter 0cap.set(cv2.CAP_PROP_POS_FRAMES, 0)if not ret:breakt6 time.time()print(t6 - t5 , t6 - t5)try:frame_queue.put_nowait(frame)except queue.Full:# 如果队列满了丢弃旧帧以防止卡顿frame_queue.get_nowait()frame_queue.put_nowait(frame)time.sleep(1.0 / fps)# 清空队列可选如果只想处理最新帧if not frame_queue.empty():frame_queue.get(blockTrue, timeout50000)# 将最新帧放入队列frame_queue.put(frame)cap.release()def read_video1(self, video_path, frame_queue):cap cv2.VideoCapture(video_path)total_frames int(cap.get(cv2.CAP_PROP_FRAME_COUNT))print(fTotal frames in the video: {total_frames})while cap.isOpened() and self.is_switchOn:#time.sleep(0.05)# print(111)t5 time.time()ret, frame cap.read()if not ret:breakt6 time.time()print(t6 - t5 , t6 - t5)try:frame_queue.put_nowait(frame)except queue.Full:# 如果队列满了丢弃旧帧以防止卡顿frame_queue.get_nowait()frame_queue.put_nowait(frame)# 清空队列可选如果只想处理最新帧if not frame_queue.empty():frame_queue.get(blockTrue, timeout50000)# 将最新帧放入队列frame_queue.put(frame)cap.release()def saveImg(self, save_queue):while self.is_switchOn:try:frame1 save_queue.get(blockTrue, timeout500) # 设置超时以避免无限等待self.ffmpegSaveVideo.stdin.write(frame1.tobytes())except queue.Empty:print(No show frame available)self.ffmpegSaveVideo.stdin.close()self.ffmpegSaveVideo.wait()except Exception as e:print(fAn error occurred: {e})breakself.ffmpegSaveVideo.stdin.close()self.ffmpegSaveVideo.wait()def pushImg(self, push_queue):print(start push images)while self.is_switchOn:try:frame1 push_queue.get(blockTrue, timeout500) # 设置超时以避免无限等待self.pipe.stdin.write(frame1.tobytes())except queue.Empty:print(No show frame available)self.pipe.stdin.close()self.pipe.wait()except Exception as e:print(fAn error occurred: {e})breakself.pipe.stdin.close()self.pipe.wait()# 图像处理函数def process_frame(self, frame_queue):while self.is_switchOn:try:# 阻塞直到队列中有帧可用t1 time.time()frame frame_queue.get(blockTrue, timeout600) # 设置超时以避免无限等待t2 time.time()# print(t2 - t1 , t2 - t1)results self.model(frame, show_conf False, verboseFalse)t3 time.time()# print(t3 - t2 , t3 - t2)for result in results:boxes result.boxes # Boxes 对象用于边界框输出# masks result.masks # Masks 对象用于分割掩码输出# keypoints result.keypoints # Keypoints 对象用于姿态输出# probs result.probs # Probs 对象用于分类输出# cv2.imshow(result,masks)frame result.plot(conf False, line_width 1, font_size 0.2) t4 time.time()print(t4 - t1 , t4 - t1)frame cv2.resize(frame, (1280,720))self.save_queue.put(frame)self.push_queue.put(frame) print(Processing frame)except queue.Empty:print(No frame available)# out.release()# self.ffmpegSaveVideo.stdin.close()# self.ffmpegSaveVideo.wait()# self.pipe.stdin.close()# self.pipe.wait()# breakexcept Exception as e:print(fAn error occurred: {e})breakdef listen_for_stop(self, listenVule):self.is_switchOn listenVuledef _listen_for_stop(self):# 监听外部状态例如键盘输入while self.is_switchOn:user_input input(输入 stop 关闭视频流: )if user_input.strip().lower() stop:self.is_switchOn Falseprint(收到关闭信号正在关闭视频流...)breakdef startThread(self, vide_path):frame_queue queue.Queue(maxsize20)# 创建并启动线程video_thread threading.Thread(targetself.read_video, args(vide_path, frame_queue))process_thread threading.Thread(targetself.process_frame, args(frame_queue,))push_thread threading.Thread(targetself.pushImg, args(self.push_queue,))save_thread threading.Thread(targetself.saveImg, args(self.save_queue,))listen_thread threading.Thread(targetself._listen_for_stop, args())#show_thread threading.Thread(targetself.dealImg, args(self.show_queue,))video_thread.start()process_thread.start()push_thread.start()save_thread.start()listen_thread.start()#show_thread.start()# 等待线程完成video_thread.join()process_thread.join()push_thread.join()save_thread.join()listen_thread.join()#show_thread.join()if __name__ __main__:model_path ./model/best640v8nLast.ptvide_path D:/datasets/chengdou/DJI_0180.MP4rtmp rtmp://192.168.100.20:1935/stream/examplegetDetect multiDealImg(model_path, rtmp)getDetect.startThread(vide_path)
