我有一个带有picam的Raspberry pi,我想将视频流传输到服务器以对其进行处理并运行对象检测软件。我通过UDP套接字将视频流发送到服务器,并接收无法转换为Mats的字节数组。
python发送者:
import socket
import time
import picamera
import sys
# msg = b'test'
addr = ("192.168.1.79", 7777)
start = time.time()
# Connect a client socket to my_server:8000 (change my_server to the
# hostname of your server)
s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
s.connect(('192.168.1.79', 7777))
con = s.makefile('wb');
print("Conectado")
# Make a file-like object out of the connection
# connection = s.makefile('wb')
try:
with picamera.PiCamera() as camera:
camera.resolution = (640, 480)
camera.framerate = 24
print("Cargando camara")
camera.start_preview()
time.sleep(2)
print("Enviando")
camera.start_recording(con, format='h264')
camera.wait_recording(10)
camera.stop_recording()
finally:
con.close()
s.close()
c#接收器:
using Emgu.CV;
using Emgu.CV.UI;
using System;
using System.Drawing;
using System.Net;
using System.Net.Sockets;
namespace EgmuCv {
class Decoder{
private const int listenPort = 7777;
public static void streamToImage(ImageBox img) {
UdpClient listener = new UdpClient(listenPort);
IPEndPoint groupEP = new IPEndPoint(IPAddress.Any, listenPort);
Mat mat = new Mat(new Size(640, 480), Emgu.CV.CvEnum.DepthType.Cv8U, 3);
try {
while (true) {
Console.WriteLine("Waiting for broadcast");
byte[] bytes = listener.Receive(ref groupEP);
// CvInvoke.Imdecode(bytes, Emgu.CV.CvEnum.ImreadModes.ReducedColor8, mat);
/** Code to turn bytes into Mat **/
// img.Image = mat;
}
} catch (SocketException e) {
Console.WriteLine(e);
} finally {
listener.Close();
}
}
}
}
答案 0 :(得分:0)
经过一些研究,我发现h.264数据段不能流传输到UDP而是TCP,并且如果要在对象检测程序中使用视频,h.264根本不是一个好主意。 一个更好的主意是使用TCP套接字发送单个MJPEG帧,虽然速度稍慢,但是却获得了640X480 @ 30的好处,这远远不够,而且转换为Mat对象的方式也更容易