Seeking Guidance on Building Real-Time Video Processing Solution for Vehicle Detection on Azure

19149916 6 Reputation points
2023-09-18T10:51:05.12+00:00

Hello Azure Community,

I'm currently working on a project for a client that involves the development of a real-time video processing solution for detecting vehicles in an area using Raspberry Pi IoT device. The project requires multiple cameras sending live video streams to Azure for processing and displaying the results on a web application in real-time.

I'm looking for guidance on the best Azure services and architecture to achieve this efficiently and effectively. Specifically, I'd like to know:

  1. Which Azure services are recommended for securely handling live video streams from Raspberry Pi devices to Azure?
  2. How can I set up real-time image processing using Azure Machine Learning Service and Azure Functions?
  3. What Azure services are suitable for hosting a web application to display real-time results?
  4. Are there any best practices for ensuring data security during transmission and storage?
  5. What monitoring and issue resolution tools or practices should I consider for this project?

I believe there are two ways to go about
one could be going with saving live streams to azure blob storage and using the Azure machine learning to deploy our ml model as service and then using azure function blob trigger to run the inference on the video stream and then saving it to another blob storage to display the stream on a web app (Not sure should we use iot hub or media service)

OR

Using Azure video analytics and deploy our custom model their to run inference and then display resulted live stream on a web app

I appreciate any insights, tips, or experiences you can share to help me plan and execute this project successfully. Thank you in advance for your assistance.

Azure AI Video Indexer
Azure AI Video Indexer
An Azure video analytics service that uses AI to extract actionable insights from stored videos.
47 questions
Azure Media Services
Azure Media Services
A group of Azure services that includes encoding, format conversion, on-demand streaming, content protection, and live streaming services.
305 questions
Azure Blob Storage
Azure Blob Storage
An Azure service that stores unstructured data in the cloud as blobs.
2,449 questions
Azure IoT Hub
Azure IoT Hub
An Azure service that enables bidirectional communication between internet of things (IoT) devices and applications.
1,128 questions

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. AshokPeddakotla-MSFT 27,966 Reputation points
    2023-09-20T08:28:18.1733333+00:00

    19149916 Greetings!

    It sounds like you have an interesting project in the works. I'd be happy to help you with your questions.

    In addition to romungi-MSFT suggestions, please see below information as well and let us know if you have any further queries.

    Which Azure services are recommended for securely handling live video streams from Raspberry Pi devices to Azure?

    Azure IoT Hub and Azure Media Services are recommended for securely handling live video streams from Raspberry Pi devices to Azure.

    Azure IoT Hub is a managed service that enables secure and reliable bi-directional communication between IoT devices and the cloud. It provides a secure and scalable way to connect, provision, and manage billions of IoT devices sending and receiving trillions of messages per month.

    Azure Media Services is a cloud-based media workflow platform that enables you to build scalable, cost-effective, and secure media solutions. It provides a range of capabilities for encoding, streaming, and delivering media content to a wide range of devices.

    You can use Azure IoT Hub to securely connect your Raspberry Pi devices to the cloud, and then use Azure Media Services to process and stream the live video feeds from those devices.

    How can I set up real-time image processing using Azure Machine Learning Service and Azure Functions?

    To set up real-time image processing using Azure Machine Learning Service and Azure Functions, you can use Azure Functions to trigger the image processing pipeline in Azure Machine Learning Service. You can use Azure Blob Storage to store the images and Azure Functions to trigger the image processing pipeline whenever a new image is uploaded to the storage account.

    Please see Apply machine learning models in Azure Functions with Python and TensorFlow for more details.

    What Azure services are suitable for hosting a web application to display real-time results?

    For hosting a web application to display real-time results, you can use Azure App Service or Azure Kubernetes Service (AKS). Azure App Service is a fully managed platform for building, deploying, and scaling web apps, while AKS is a fully managed Kubernetes container orchestration service that can be used to deploy and manage containerized applications. You can also use Azure SignalR Service to enable real-time communication between the web application and the backend services.

    Are there any best practices for ensuring data security during transmission and storage?

    To ensure data security during transmission and storage, you can use Azure Security Center to monitor and manage security across your Azure resources. You can also use Azure Key Vault to securely store and manage cryptographic keys and secrets used by your applications.

    What monitoring and issue resolution tools or practices should I consider for this project?

    For monitoring and issue resolution, you can use Azure Monitor to collect and analyze telemetry data from your Azure resources and applications.

    You can also use Azure Application Insights to monitor the performance and usage of your web application. Additionally, you can set up alerts and notifications using Azure Monitor to proactively identify and resolve issues.

    Do let us know if that helps or have any other queries.

    If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further queries do let us know.