问题描述
我有一个应用程序，它要求我使用 RTSP 流，从视频中提取帧，对视频帧执行一些处理，对帧进行注释，然后将视频重新传输为新的 RTSP 流。听起来很简单。
import gi
import cv2
import os
# import required library like Gstreamer and GstreamerRtspServer
gi.require_version('Gst','1.0')
gi.require_version('GstRtspServer','1.0')
from gi.repository import Gst,GstRtspServer,GObject
# Sensor Factory class which inherits the GstRtspServer base class and add
# properties to it.
class SensorFactory(GstRtspServer.RTSPMediaFactory):
def __init__(self,**properties):
super(SensorFactory,self).__init__(**properties)
# Create a connection to our input RTSP stream and obtain the width and height
self.cap = cv2.VideoCapture("rtspsrc location=rtsp://root:admin@192.168.88.248/stream1 ! decodebin ! videoconvert ! appsink max-buffers=3 drop=true")
width = int(self.cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(self.cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
print(width)
print(height)
self.number_frames = 0
self.fps = 30
self.duration = 1 / self.fps * Gst.SECOND # duration of a frame in nanoseconds
self.launch_string = 'appsrc name=source is-live=true block=true format=GST_FORMAT_TIME ' \
'caps=video/x-raw,format=BGR,width={},height={},framerate={}/1 ' \
'! videoconvert ! video/x-raw,format=I420 ' \
'! x264enc speed-preset=ultrafast tune=zerolatency ' \
'! rtph264pay config-interval=1 name=pay0 pt=96' \
.format(width,height,self.fps)
# Method for grabbing frames from the video capture,process,then pushing annotated images to streaming buffer
def on_need_data(self,src,lenght):
if self.cap.isOpened():
ret,frame = self.cap.read()
if ret:
# --------------------------
# --------------------------
# Do processing here
# --------------------------
# --------------------------
data = frame.tobytes()
buf = Gst.Buffer.new_allocate(None,len(data),None)
buf.fill(0,data)
buf.duration = self.duration
timestamp = self.number_frames * self.duration
buf.pts = buf.dts = int(timestamp)
buf.offset = timestamp
self.number_frames += 1
retval = src.emit('push-buffer',buf)
print('pushed buffer,frame {},duration {} ns,durations {} s'.format(self.number_frames,self.duration,self.duration / Gst.SECOND))
if retval != Gst.FlowReturn.OK:
print(retval)
# attach the launch string to the override method
def do_create_element(self,url):
return Gst.parse_launch(self.launch_string)
# attaching the source element to the rtsp media
def do_configure(self,rtsp_media):
self.number_frames = 0
appsrc = rtsp_media.get_element().get_child_by_name('source')
appsrc.connect('need-data',self.on_need_data)
# Rtsp server implementation where we attach the factory sensor with the stream uri
class GstServer(GstRtspServer.RTSPServer):
def __init__(self,**properties):
super(GstServer,self).__init__(**properties)
self.factory = SensorFactory()
self.factory.set_shared(True)
self.get_mount_points().add_factory("/my_stream",self.factory)
self.attach(None)
# initializing the threads and running the stream on loop.
GObject.threads_init()
Gst.init(None)
server = GstServer()
loop = GObject.MainLoop()
loop.run()
# To connect:
# rtsp://localhost:8554/my_stream
我在标记为：
的部分中执行我的处理 # --------------------------
# --------------------------
# Do processing here
# --------------------------
# --------------------------
现在这段代码非常适合处理时间恒定的情况，因为那时我知道输出 rtsp 流 fps。 但是，我的处理时间不固定；花费的时间取决于来自源 rtsp 流的输入帧。就上下文而言，我正在构建一个人脸识别应用程序，帧中每个检测到的人脸的处理时间约为 50 毫秒 + 100 毫秒。
如何使输出 rtsp 流以这种可变帧速率工作？
我的猜测是我需要修改 gstreamer 管道字符串：
self.launch_string = 'appsrc name=source is-live=true block=true format=GST_FORMAT_TIME ' \
'caps=video/x-raw,self.fps)
而且我可能也需要更改这部分：
buf.duration = self.duration
timestamp = self.number_frames * self.duration
buf.pts = buf.dts = int(timestamp)
buf.offset = timestamp
解决方法
暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！
如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。
小编邮箱:dio#foxmail.com (将#修改为@）