Video

warning

The document is a continuation of the previous document, if you have landed directly on this page then, Please read from page Get started.

What is Video Component?

List of features implemented in Shunya stack

With Shunya stack Video component you can

1. Capture video frames from these sources.
   1. Camera (USB/Embedded)
   2. Video file.
   3. RTSP Stream (CCTV stream)
2. Store in 2 formats
   1. Image (Screenshot of one frame)
   2. Video

Using Video component

Requirements to use Video component

1. Shunya OS installed (supported Arm devices) or Shunya OS docker container (X86 based windows/linux devices)
2. Availability of any one video source. (for example: Webcamera or Video file)

Steps to use Video component

1. Set Capture settings in Shunya.
2. Set Store settings in Shunya.
3. Implementation for Capture
4. Implementation for Store

note

Run the steps given below inside Shunya OS installed (supported Arm devices) or Shunya OS docker container (X86 based windows/linux devices) terminals.

Step 1: Set Capture settings in Shunya.

Set Capture source, by editing the config file /etc/shunya/config.json inside Shunya OS.

A simple configuration should contain these parameters.

Config	Description
cameraId	Camera device number (optional)
videoPath	Full Path to the Video file (optional)
streamUrl	Url for capturing RTSP/UDP streams. (optional)

Example configuration for Video source

Camera

Writing the below json in `/etc/shunya/config.json` file, will tell the Shunya stack to capture from a webcam or embedded camera

{ 
    "video-source": {
        "cameraId": "0"
    }
}

Video file

Writing the below json in `/etc/shunya/config.json` file, will tell the Shunya stack to capture from video (eg: 15.mp4)

{
    "video-source": {
        "videoPath": "/home/shunya/15.mp4"
    }
}

Video stream

Writing the below json in `/etc/shunya/config.json` file, will tell the Shunya stack to capture from udp or rtsp stream

{
    "video-source": {
        "streamUrl": "udp://localhost:9090"
    }
}

Pro-tip

Lets say your video source changed from video file to webcam, then you dont have to change the whole code, you just need to change the configuration and the video component will adjust accordingly.

Step 2: Set Store settings in Shunya.

Set Store Location, by editing the config file /etc/shunya/config.json inside Shunya OS.

A simple configuration should contain these parameters.

Config	Description
storeType	Storing format, takes values image or video
storePath	Full Path to the folder you want to store.
videoFps	FPS of the output video (only for storeType = video)

Example configuration for Video store

You can store both images as well as video

Image

Writing the below json in /etc/shunya/config.json file, will tell the Shunya stack to store an image at /home/shunya/Desktop folder

"imageStore": {
    "storeType": "image",
    "storePath": "/home/shunya/Desktop"
}

Video file

Writing the below json in /etc/shunya/config.json file, will tell the Shunya stack to store an video at /home/shunya/Desktop folder

"videoStore": {
    "storeType": "video",
    "storePath": "/home/shunya/Desktop",
    "videoFps": 10  // FPS of the video stored
}

Pro-tip

Lets say your video store path has changed, then you dont have to change the whole code ore re-compile, you just need to change the configuration and the video component will adjust accordingly.

Step 3: Implementation for Capture

Let's try to implement video Capture using Shunya stack

Steps are :

1. Start with an ready to use template

   git clone https://gitlab.iotiot.in/repo-public/examples.git
   cd examples/simple-examples/capture-video

2. Open the capture-video.cpp in an text editor and modify as per your use case.

3. First we load the JSON settings for Capture, add the code in your main() function

   captureObj src;
   src = newCaptureDevice("video-source"); /* Argument = JSON Title, Load settings from JSON file */

4. To Capture one frame, modify the code

   cv::Mat frame  = captureFrameToMem(&src); /* Capture one frame at a time in a loop*/
   if (frame.empty()){ /* Check if frame is empty*/
       fprintf(stderr, "Frame empty.");
       closeCapture(&src); /* Close Video source and exit */
       return 0;
   }
   /* frame variable contains one frame, this can be sent to other components for processing */
   

5. Once you are done editing, save and compile the code , by running

   mkdir build && cd build
   cmake ../
   make 

6. Run the code

   sudo ./capture-video

Step 4: Implementation for Store.

Let's try to implement video Store using Shunya stack

Steps are :

1. Start with an ready to use template

   git clone https://gitlab.iotiot.in/repo-public/examples.git
   cd examples/simple-examples/store-video

2. Open the store-video.cpp in an text editor and modify as per your use case.

3. First we load the JSON settings for Store, add the code in your main() function

   Image

   storeObj imageStore; /* Image Store instance */
   imageStore = newImageStore("imageStore"); /* Argument = JSON Title, Load settings from JSON file */

   Video file

   storeObj videoStore;  /* Video Store instance */
   videoStore = newVideoStore("videoStore"); /* Argument = JSON Title, Load settings from JSON file */

4. To Store one frame, modify the code

   Image

   storeFrameToDisk(&imageStore, frame, "file.jpg"); /* Store frame as file.jpg*/

   Video file

   storeStreamToDisk(&videoStore, frame, "file.avi"); /* Store frame into file.avi*/

5. Cleanup the store operations with,

   Image

   closeStore(&imageStore);

   Video file

   closeStore(&videoStore);

6. Once you are done editing, save and compile the code , by running

   mkdir build && cd build
   cmake ../
   make 

7. Run the code

   sudo ./store-video

Facing errors with the component?