Security Camera HTTP Image download

As part of a contract a customer sent me a Uniview IPC3635SB-ADZK-I0 Security camera for a proof of concept(PoC) project. Before the PoC I wanted to explore the camera functionality in more depth, especially how to retrieve individual images from the camera, remotely control it’s zoom, focus, pan, tilt etc.. I’m trying to source a couple of other vendors’ security cameras with remotely controllable pan and tilt for testing.

Uniview IPC3635SB-ADZK-I0 Security camera

It appears that many cameras support retrieving the latest image with an HyperText Transfer Protocol (HTTP) GET so that looked like a good place to start. For the next couple of posts the camera will be sitting on the bookcase in my office looking through the window at the backyard.

Unv camera software live view of my backyard

One thing I did notice (then confirmed with Telerik Fiddler and in the camera configuration) was that the camera was configured to use Digest authentication(RFC 2069) which broke my initial attempt with a Universal Windows Platform(UWP) application.

Telerik Fiddler showing 401 authorisation challenge

My .Net Core 5 console application is as simple possible, it just downloads an image from the camera “snapshot” endpoint (In this case http://10.0.0.47:85/images/snapshot.jpg) and saves it to the local filesystem.

class Program
{
	static async Task Main(string[] args)
	{
		await Parser.Default.ParseArguments<CommandLineOptions>(args)
			.WithNotParsed(HandleParseError)
			.WithParsedAsync(ApplicationCore);
	}

	private static async Task ApplicationCore(CommandLineOptions options)
	{
		Console.WriteLine($"Camera:{options.CameraUrl} UserName:{options.UserName} filename:{options.Filename}");

		using (var client = new WebClient())
		{
			NetworkCredential networkCredential = new NetworkCredential()
			{
				UserName = options.UserName,
				Password = options.Password
			};

			client.Credentials = networkCredential;

			try
			{
				await client.DownloadFileTaskAsync(new Uri(options.CameraUrl), options.Filename);
			}
			catch (Exception ex)
			{
				Console.WriteLine($"File download failed {ex.Message}");
			}
		}

		Console.WriteLine("Press <enter> to exit");
		Console.ReadLine();
	}

	private static void HandleParseError(IEnumerable<Error> errors)
	{
		if (errors.IsVersion())
		{
			Console.WriteLine("Version Request");
			return;
		}

		if (errors.IsHelp())
		{
			Console.WriteLine("Help Request");
			return;
		}
		Console.WriteLine("Parser Fail");
	}
}

After confirming the program was working I used the excellent RaspberryDebugger to download the application and debug it on a Raspberry PI 3 running the Raspberry PI OS.

Visual Studio 2019 Debug Output showing application download process

Once the application had finished running on the device I wanted to check that the file was on the local filesystem. I used Putty to connect to the Raspberry PI then searched for LatestImage.jpg.

Linux find utility displaying the location of the downloaded file

I though about using a utility like scp to download the image file but decided (because I have been using Microsoft Window since WIndows 286) to install xrdp an open-source Remote Desktop Protocol(RDP) server so I could use a Windows 10 RDP client.

xrdp login screen
xrdp home screen
xrdp file manager display files in application deployment directory
Raspberry PI OS default image view

Now that the basics are working my plan is to figure out how to control the camera, display live video with the Real Time Streaming Protocol(RTSP) upload images to Azure Cognitive Services for processing and use ML.Net to process them locally.

This post was about selecting the tooling I’m comfortable with and configuring my development environment so they work well together. The next step will be using Open Network Video Interface Forum (ONVIF) to discover, determine the capabilities of and then control the camera (for this device just zoom and focus).