---
title: Amazon Rekognition
description: |-
  Amazon Rekognition makes it easy to add image and video analysis to
          your applications.
breadcrumbs: Docs > Integrations > Amazon Rekognition
---

# Amazon Rekognition
Integration version1.0.0
## Overview{% #overview %}

Amazon Rekognition makes it easy to add image and video analysis to your applications. You just provide an image or video to the Rekognition API, and the service can identify objects, people, text, scenes, and activities.

Enable this integration to see all your Rekognition metrics in Datadog.

## Setup{% #setup %}

### Installation{% #installation %}

If you haven't already, set up the [Amazon Web Services integration](https://docs.datadoghq.com/integrations/amazon_web_services/) first.

### Metric collection{% #metric-collection %}

1. In the [AWS integration page](https://app.datadoghq.com/integrations/amazon-web-services), ensure that `Rekognition` is enabled under the `Metric Collection` tab.
1. Install the [Datadog - Amazon Rekognition integration](https://app.datadoghq.com/integrations/amazon-rekognition).

### Log collection{% #log-collection %}

#### Enable logging{% #enable-logging %}

Configure Amazon Rekognition to send logs either to a S3 bucket or to CloudWatch.

**Note**: If you log to a S3 bucket, make sure that `amazon_rekognition` is set as *Target prefix*.

#### Send logs to Datadog{% #send-logs-to-datadog %}

1. If you haven't already, set up the [Datadog Forwarder Lambda function](https://docs.datadoghq.com/logs/guide/forwarder/).

1. Once the Lambda function is installed, manually add a trigger on the S3 bucket or CloudWatch log group that contains your Amazon Rekognition logs in the AWS console:

   - [Add a manual trigger on the S3 bucket](https://docs.datadoghq.com/logs/guide/send-aws-services-logs-with-the-datadog-lambda-function/#collecting-logs-from-s3-buckets)
   - [Add a manual trigger on the CloudWatch Log Group](https://docs.datadoghq.com/logs/guide/send-aws-services-logs-with-the-datadog-lambda-function/#collecting-logs-from-cloudwatch-log-group)

## Data Collected{% #data-collected %}

### Metrics{% #metrics %}

|  |
|  |
| **aws.rekognition.detected\_face\_count**(count)          | The average number of faces detected with the IndexFaces or DetectFaces operation.                                                                                          |
| **aws.rekognition.detected\_face\_count.sum**(count)      | The sum of the number of faces detected with the IndexFaces or DetectFaces operation.                                                                                       |
| **aws.rekognition.response\_time**(count)                 | The time in milliseconds for Rekognition to compute the response.*Shown as millisecond*                                                                                     |
| **aws.rekognition.server\_error\_count**(count)           | The number of server errors.                                                                                                                                                |
| **aws.rekognition.server\_error\_count.sum**(count)       | The sum of the number of server errors.                                                                                                                                     |
| **aws.rekognition.successful\_request\_count**(count)     | The average number of successful requests.                                                                                                                                  |
| **aws.rekognition.successful\_request\_count.sum**(count) | The sum of the number of successful requests.                                                                                                                               |
| **aws.rekognition.throttled\_count**(count)               | The average number of throttled requests. Rekognition throttles a request when it receives more requests than the limit of transactions per second set for your account.    |
| **aws.rekognition.throttled\_count.sum**(count)           | The sum of the number of throttled requests. Rekognition throttles a request when it receives more requests than the limit of transactions per second set for your account. |
| **aws.rekognition.user\_error\_count**(count)             | The average number of user errors (invalid parameters - invalid image - no permission etc).                                                                                 |
| **aws.rekognition.user\_error\_count.sum**(count)         | The sum of the number of user errors (invalid parameters - invalid image - no permission etc).                                                                              |
| **aws.rekognition.deteceted\_label\_count**(count)        | The average number of labels detected with the DetectLabels operation.                                                                                                      |
| **aws.rekognition.deteceted\_label\_count.sum**(count)    | The sum of the number of labels detected with the DetectLabels operation.                                                                                                   |
| **aws.rekognition.response\_time.data\_samples**(count)   | The time in milliseconds for Rekognition to compute the response.                                                                                                           |

### Events{% #events %}

The Amazon Rekognition integration does not include any events.

### Service Checks{% #service-checks %}

The Amazon Rekognition integration does not include any service checks.

## Troubleshooting{% #troubleshooting %}

Need help? Contact [Datadog support](https://docs.datadoghq.com/help/).