amazon-kinesis-video-streams-producer-sdk-cpp/raspberry-pi.md at master · awslabs/amazon-kinesis-video-streams-producer-sdk-cpp - Chuyên Trang Chia Sẻ Kiến Thức Thời Trang Mới Nhất

The following steps were tested on a Debian buster platform

Mục lục

Installing libraries needed to build

Run the following commands to install the prerequisite libraries to get started:

sudo apt-get install cmake m4 git build-essential
sudo apt-get install gstreamer1.0-plugins-good gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly gstreamer1.0-tools

Also, install the gstreamer1.0-omx package to get the omxh264enc hardware encoder:

sudo apt-get install gstreamer1.0-omx

If building JNI, ensure Java JDK is available on the system. To check this, run:
java -showversion
If nothing shows up, install OpenJDK:
sudo apt-get install openjdk-8-jdk

How to run sample applications for sending media to KVS using GStreamer:

Ensure you run cmake with BUILD_GSTREAMER_PLUGIN=ON option from build directory

Setting credentials in environment variables

Define AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY environment variables with the AWS access key id and secret key:

$ export AWS_ACCESS_KEY_ID=YourAccessKeyId
$ export AWS_SECRET_ACCESS_KEY=YourSecretAccessKey

optionally, set AWS_SESSION_TOKEN if integrating with temporary token and AWS_DEFAULT_REGION for the region other than us-west-2

Set GST_PLUGIN_PATH in environment variables

$ export GST_PLUGIN_PATH=>SDK_FOLDER_PATH/build

Discovering audio and video devices available in your system.

Run the gst-device-monitor-1.0 command to identify available media devices in your system. An example output as follows:

Probing devices...


Device found:

	name  : H264 USB Camera: USB Camera
	class : Video/Source
	caps  : video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)1920, height=(int)1080, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)1920, height=(int)1080, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)1280, height=(int)720, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)800, height=(int)600, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)640, height=(int)480, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)640, height=(int)360, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)352, height=(int)288, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	        video/x-h264, stream-format=(string)byte-stream, alignment=(string)au, width=(int)320, height=(int)240, pixel-aspect-ratio=(fraction)1/1, colorimetry=(string){ 2:4:7:1 }, framerate=(fraction){ 30/1, 25/1, 15/1 };
	properties:
		device.path = /dev/video4
		udev-probed = false
		device.api = v4l2
		v4l2.device.driver = uvcvideo
		v4l2.device.card = "H264\ USB\ Camera:\ USB\ Camera"
		v4l2.device.bus_info = usb-3f980000.usb-1.3
		v4l2.device.version = 265767 (0x00040e27)
		v4l2.device.capabilities = 2216689665 (0x84200001)
		v4l2.device.device_caps = 69206017 (0x04200001)
	gst-launch-1.0 v4l2src device=/dev/video4 ! ...

Running the `gst-launch-1.0` command to start streaming from a RTSP camera source.

$ gst-launch-1.0 -v rtspsrc location=rtsp://YourCameraRtspUrl short-header=TRUE ! rtph264depay ! h264parse ! kvssink stream-name=YourStreamName storage-size=128

Note: If you are using IoT credentials then you can pass them as parameters to the gst-launch-1.0 command

$ gst-launch-1.0 -v rtspsrc location="rtsp://YourCameraRtspUrl" short-header=TRUE ! rtph264depay ! h264parse ! kvssink stream-name="iot-stream" iot-certificate="iot-certificate,endpoint=endpoint,cert-path=/path/to/certificate,key-path=/path/to/private/key,ca-path=/path/to/ca-cert,role-aliases=role-aliases"

You can find the RTSP URL from your IP camera manual or manufacturers product page.

Running the `gst-launch-1.0` command to start streaming from USB camera source which has h264 encoded stream already:

$ gst-launch-1.0 -v v4l2src device=/dev/video0 ! h264parse ! video/x-h264,stream-format=avc,alignment=au ! kvssink stream-name=YourStreamName storage-size=128 access-key="YourAccessKey" secret-key="YourSecretKey"

Running the `gst-launch-1.0` command to start streaming from camera source:

$ gst-launch-1.0 -v v4l2src do-timestamp=TRUE device=/dev/video0 ! videoconvert ! video/x-raw,format=I420,width=640,height=480,framerate=30/1 ! omxh264enc periodicty-idr=45 inline-header=FALSE ! h264parse ! video/x-h264,stream-format=avc,alignment=au ! kvssink stream-name=YourStreamName access-key="YourAccessKey" secret-key="YourSecretKey"

or use a different encoder

$ gst-launch-1.0 -v v4l2src device=/dev/video0 ! videoconvert ! video/x-raw,format=I420,width=640,height=480,framerate=30/1 ! x264enc  bframes=0 key-int-max=45 bitrate=500 tune=zerolatency ! video/x-h264,stream-format=avc,alignment=au ! kvssink stream-name=YourStreamName storage-size=128 access-key="YourAccessKey" secret-key="YourSecretKey"

Running the `gst-launch-1.0` command to start streaming both audio and video:

Please ensure that audio drivers are installed first by running

apt-get install libasound2-dev

then you can use the following following command to find the capture card and device number.

arecord -l (or arecord --list-devices)

the output should look like the following:

**** List of CAPTURE Hardware Devices ****
card 2: U0x46d0x825 [USB Device 0x46d:0x825], device 0: USB Audio [USB Audio]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 3: Camera [H264 USB Camera], device 0: USB Audio [USB Audio]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

The audio recording device is represented by hw:card_number,device_numer. So to use the second device in the example, use hw:3,0 as the device in gst-launch-1.0 command.

gst-launch-1.0 -v v4l2src device=/dev/video0 ! videoconvert ! video/x-raw,width=640,height=480,framerate=30/1,format=I420 ! omxh264enc periodicty-idr=45 inline-header=FALSE ! h264parse ! video/x-h264,stream-format=avc,alignment=au,profile=baseline ! kvssink name=sink stream-name="my-stream-name" access-key="YourAccessKey" secret-key="YourSecretKey" alsasrc device=hw:1,0 ! audioconvert ! avenc_aac ! queue ! sink.

if your camera supports outputting h264 encoded stream directly, then you can use this command:

gst-launch-1.0 -v v4l2src device=/dev/video0 ! h264parse ! video/x-h264,stream-format=avc,alignment=au ! kvssink name=sink stream-name="my-stream-name" access-key="YourAccessKey" secret-key="YourSecretKey" alsasrc device=hw:1,0 ! audioconvert ! avenc_aac ! queue ! sink.

Running the GStreamer webcam sample application

The sample application kvs_gstreamer_sample in the build directory uses GStreamer pipeline to get video data from the camera. Launch it with a stream name and it will start streaming from the camera. The user can also supply a streaming resolution (width and height) through command line arguments.

Usage: AWS_ACCESS_KEY_ID=YourAccessKeyId AWS_SECRET_ACCESS_KEY=YourSecretAccessKey ./kvs_gstreamer_sample <my_stream_name> -w <width> -h <height> -f <framerate> -b <bitrateInKBPS>

A. If resolution is provided then the sample will try to check if the camera supports that resolution. If it does detect that the camera can support the resolution supplied in command line, then streaming starts; else, it will fail with an error message Resolution not supported.
B. If no resolution is specified, the sample application will try to use these three resolutions 640×480, 1280×720 and 1920×1080 and will start streaming once the camera supported resolution is detected.

Running the GStreamer RTSP sample application

kvs_gstreamer_sample supports sending video from a RTSP URL (IP camera). You can find the RTSP URL from your IP camera manual or manufacturers product page. Change your current working direcctory to build directory. Launch it with a stream name and rtsp_url and it will start streaming.

AWS_ACCESS_KEY_ID=YourAccessKeyId AWS_SECRET_ACCESS_KEY=YourSecretAccessKey ./kvs_gstreamer_sample <my-rtsp-stream> <my_rtsp_url>

Running the GStreamer sample application to upload a video file

kvs_gstreamer_sample supports uploading a video that is either MKV, MPEGTS, or MP4. The sample application expects the video is encoded in H264.

Change your current working directory to build. Launch the sample application with a stream name and a path to the file and it will start streaming.

AWS_ACCESS_KEY_ID=YourAccessKeyId AWS_SECRET_ACCESS_KEY=YourSecretAccessKey ./kvs_gstreamer_sample <my-stream> </path/to/file>

Running the `gst-launch-1.0` command to upload MKV file that contains both audio and video in Raspberry-PI. Note that video should be H264 encoded and audio should be AAC encoded.

gst-launch-1.0 -v filesrc location="YourAudioVideo.mkv" ! matroskademux name=demux ! queue ! h264parse ! kvssink name=sink stream-name="my_stream_name" access-key="YourAccessKeyId" secret-key="YourSecretAccessKey" streaming-type=offline demux. ! queue ! aacparse ! sink.

Running the `gst-launch-1.0` command to upload MP4 file that contains both audio and video:

gst-launch-1.0 -v  filesrc location="YourAudioVideo.mp4" ! qtdemux name=demux ! queue ! h264parse ! video/x-h264,stream-format=avc,alignment=au ! kvssink name=sink stream-name="audio-video-file" access-key="YourAccessKeyId" secret-key="YourSecretAccessKey" streaming-type=offline demux. ! queue ! aacparse ! sink.

Running the `gst-launch-1.0` command to upload MPEG2TS file that contains both audio and video:

gst-launch-1.0 -v  filesrc location="YourAudioVideo.ts" ! tsdemux name=demux ! queue ! h264parse ! video/x-h264,stream-format=avc,alignment=au ! kvssink name=sink stream-name="audio-video-file" access-key="YourAccessKeyId" secret-key="YourSecretAccessKey" streaming-type=offline demux. ! queue ! aacparse ! sink.

Running the `gst-launch-1.0` command with Iot-certificate and different stream-names than the thing-name

Note: Supply a the matching iot-thing-name (that the certificate points to) and we can stream to multiple stream-names (without the stream-name needing to be the same as the thing-name) using the same certificate credentials. iot-thing-name and stream-name can be completely different as long as there is a policy that allows the thing to write to the kinesis stream

$ gst-launch-1.0 -v rtspsrc location="rtsp://YourCameraRtspUrl" short-header=TRUE ! rtph264depay ! video/x-h264, format=avc,alignment=au !
 h264parse ! kvssink name=aname storage-size=512 iot-certificate="iot-certificate,endpoint=xxxxx.credentials.iot.ap-southeast-2.amazonaws.com,cert-path=/greengrass/v2/thingCert.crt,key-path=/greengrass/v2/privKey.key,ca-path=/greengrass/v2/rootCA.pem,role-aliases=KvsCameraIoTRoleAlias,iot-thing-name=myThingName123" aws-region="ap-southeast-2" log-config="/etc/mtdata/kvssink-log.config" stream-name=myThingName123-video1

Running the GStreamer sample application to upload a audio and video file

kinesis_video_gstreamer_audio_video_sample_app supports uploading a video that is either MKV, MPEGTS, or MP4. The sample application expects the video is encoded in H264 and audio is encoded in AAC format. Note: If your media uses a different format, then you can revise the pipeline elements in the sample application to suit your media format.

Change your current working directory to build. Launch the sample application with a stream name and a path to the file and it will start streaming.

AWS_ACCESS_KEY_ID=YourAccessKeyId AWS_SECRET_ACCESS_KEY=YourSecretAccessKey ./kvs_gstreamer_audio_video_sample <my-stream> </path/to/file>

Running the GStreamer sample application to stream audio and video from live source

kvs_gstreamer_audio_video_sample supports streaming audio and video from live sources such as a audio enabled webcam. First you need to figure out what your audio device is using the steps mentioned above and export it as environment variable like such:

export AWS_KVS_AUDIO_DEVICE=hw:1,0

You can also choose to use other video devices by doing

export AWS_KVS_VIDEO_DEVICE=/dev/video1

If no AWS_KVS_VIDEO_DEVICE environment variable was detected, the sample application will use the default video device.
After the environment variables are set, launch the sample application with a stream name and it will start streaming.

AWS_ACCESS_KEY_ID=YourAccessKeyId AWS_SECRET_ACCESS_KEY=YourSecretAccessKey ./kvs_gstreamer_audio_video_sample <my-stream>

Additional examples

For additional examples on using Kinesis Video Streams Java SDK and Kinesis Video Streams Parsing Library refer:

Running C++ unit tests

Note: Please set the credentials before running the unit tests:

$ export AWS_ACCESS_KEY_ID=YourAccessKeyId
$ export AWS_SECRET_ACCESS_KEY=YourSecretAccessKey
optionally, set AWS_SESSION_TOKEN if integrating with temporary token and AWS_DEFAULT_REGION for the region other than us-west-2

The executable for unit tests will be built as ./tst/producer_test inside the build directory. Launch it and it will run the unit test and kick off dummy frame streaming.

Running GStreamer Unit tests

Note: Please set the credentials before running the unit tests:

$ export AWS_ACCESS_KEY_ID=YourAccessKeyId
$ export AWS_SECRET_ACCESS_KEY=YourSecretAccessKey
optionally, set AWS_SESSION_TOKEN if integrating with temporary token and AWS_DEFAULT_REGION for the region other than us-west-2

The executable for GStreamer unit tests will be built as ./tst/gstkvsplugintest inside the build directory. Launch it and it will run the unit test and kick off dummy frame streaming.

How to configure logging for producer SDK sample applications.

For the sample demo applications included in the producer SDK (CPP), the log configuration is referred from the file kvs_log_configuration (within the samples folder).

Refer sample configuration in the folder build for details on how to set the log level (DEBUG or INFO) and output options (whether to send log output to either console or file (or both)).

Log output messages to console:
By default, the log configuration log4cplus.rootLogger=DEBUG, KvsConsoleAppender creates console appender (KvsConsoleAppender) which outputs the log messages in the console.
Log output messages to file: By adding file appender (KvsFileAppender) in the rootLogger of log4cplus as log4cplus.rootLogger=DEBUG, KvsConsoleAppender, KvsFileAppender the debug messages will be stored in kvs.log file in the sub-folder log within build directory. The filename for the logs and the location can be modified by changing the line log4cplus.appender.KvsFileAppender.File=./log/kvs.log

How to enable saving c producer log into files.

By default C producer prints all logging information to stdout.

To send log information to a file (named kvsProducerLog.index), you need to use the addFileLoggerPlatformCallbacksProvider API after ClientCallbacks has been initialized.

The addFileLoggerPlatformCallbacksProvider API takes five parameters.

First parameter is the PClientCallbacks that is created during the createCallback provider API (e.g.createDefaultCallbacksProviderWithAuthCallbacks.
Second parameter is the size of string buffer that file logger will use. Logs are buffered in the string buffer and flushed into files when the buffer is full.
Third parameter is the maximum number of files that file logger will generate. When the limit is reached, oldest log file will be deleted before creating the new one.
Fourth parameter is the absolute directory path to store the log file.
Fifth parameter uses boolean true or false and is used to allow printing logs to both stdout and a file (useful in debugging).

Open Source Dependencies

The projects depend on the following open source components. Running CMake will download and build the necessary components automatically.

Producer SDK Core

openssl (crypto and ssl) – License
curl lib – Copyright
log4cplus – License
jsoncpp – License

Unit Tests

GStreamer Demo App

gstreamer – License
gst-plugins-base
gst-plugins-good
gst-plugins-bad
gst-plugins-ugly
x264

Raspberry PI failure to load the camera device.

To check this is the case run ls /dev/video* – it should be file not found. The remedy is to run the following:

$ls /dev/video*
{not found}

$vcgencmd get_camera

Example output:

supported=1 detected=1

if the driver does not detect the camera then

Check for the camera setup and whether it’s connected properly
Run firmware update $ sudo rpi-update and restart

$sudo modprobe bcm2835-v4l2

$ls /dev/video*
{lists the device}

Raspberry PI timestamp/range assertion at runtime.

Update the Raspberry PI firmware.

$ sudo rpi-update
$ sudo reboot

Raspberry PI GStreamer assertion on gst_value_set_fraction_range_full: assertion ‘gst_util_fraction_compare (numerator_start, denominator_start, numerator_end, denominator_end) < 0’ failed. The uv4l service running in the background. Kill the service and restart the sample app.

amazon-kinesis-video-streams-producer-sdk-cpp/raspberry-pi.md at master · awslabs/amazon-kinesis-video-streams-producer-sdk-cpp