# Anomaly Detection

DeviceOn Anomaly Detection Service (ADS) is one of the services that combines anomaly detection algorithm and DeviceOn function. It not only gives customers the high accuracy identification of the error message when advertising is getting interruption but also provides the IoT device remote monitoring and management. Leverage with ***Azure Custom Vision*** to continually train the algorithm in order to overcome various errors pop-up under real field.

{% hint style="warning" %}
Before to detect your anomaly screen, please make sure your device is logged in to capture screen status.
{% endhint %}

Click the  icon to enter your API server **URL** and **API Key**.

<figure><img src="https://i.imgur.com/6aotHLj.png" alt=""><figcaption></figcaption></figure>

* **API URL & Key**: Please contact us for AI machine URL and Key, otherwise, deploy total package from Azure Marketplace.
* **Maximum Retention Days**: Maximum retention days for the warning images.
* **Interval**: ***Minimal interval*** to detect devices screen.
* **Pop-up Window**: Enable to detect popup window on the devices.
* **Screen Frozen**: Enable to detect freezing window on the devices.

Second, create a rule for your device and define the **interval** that to capture the image for inference.

<figure><img src="https://i.imgur.com/KTxdsao.png" alt=""><figcaption></figcaption></figure>

If a screen freeze or any pop-up window is detected, it will be displayed here and dynamically marked with errors

<figure><img src="https://i.imgur.com/9cS8TxE.png" alt=""><figcaption></figcaption></figure>

You can also view the screen changes of these abnormal devices through historical records.

<figure><img src="https://i.imgur.com/e7oUkF1.png" alt=""><figcaption></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://deviceon.gitbook.io/docs/web-user-interface/device-management/anomaly-detection.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.