Use Oracle Cloud Infrastructure Vision Models in Oracle Analytics

Use pretrained Oracle Cloud Infrastructure (OCI) Vision models to analyze images and videos. This enables you to build image recognition and text recognition into your applications without machine learning (ML) or artificial intelligence (AI) expertise.

For example, when you use OCI pretrained Vision models, you can detect objects such as cars and faces in photographs and then blur the images to protect identities of the individuals.

Oracle Analytics supports these models:

• Image Classification for images.
• Face Detection for images and videos.
• Text Detection for images and videos.
• Label Detection for videos.
• Object Detection for images and videos.

Prerequisites:

• See Integrate Oracle Analytics with Oracle Cloud Infrastructure Vision.
• See Prepare Images or Videos for Analysis with an OCI Vision Model.

Note:

OCI face detection with Oracle Analytics can identify a maximum of 250 faces per image.

1. On the Oracle Analytics Home page, click Create, and then click Data Flow.
2. Select the prepared dataset that references the images or videos and click Add.
3. From the Data Flow Steps pane, double-click Apply AI Model.
4. In Select AI Model, select a model, and then click OK.
5. Use the Outputs and Parameters options to configure the model.
   
   
   Description of the illustration oci-vision-face-recognition.png
   
   
6. In Apply Model, expand Parameters and configure Input Column and Input Type.
   • Input Column - Click Select a column and select the dataset column that contains the bucket URL or URIs of the images or videos. For example, if the dataset column is named URL, select URL.
     
     
     Description of the illustration vision_parameters.png
     
     
   • Input Type:
     • If you're referencing your source images by bucket, select Buckets.
     • If you're referencing your source images individually, select Images.
     • If you're referencing your source videos individually, select Videos.
   See Parameter Options for Object Detection, Image Classification, and Text Detection Analysis Models.
7. From the Data Flow Steps pane, double-click Save Data.
8. Enter a name for the dataset and select a location for saving the dataset.
   For example, you might call the dataset 'Car Parking Analysis Results'.
9. Click Save, enter a name for the data flow, and click OK to save the data flow.
10. Click Run Data Flow to analyze the images and output the results in a new dataset.
11. On the Oracle Analytics Home page, click Data, and open the dataset that you specified in Step 8.
    Use the dataset to analyze the results. See Output Data Generated for Face Detection, Object Detection, Image Classification, and Text Detection Analysis Models.

If you have fewer than 20,000 images, you can process them in a single data flow. If you have more than 20,000 images, create a separate data flow to process each bucket (that is, using a separate dataset for each bucket), and use a Sequence to sequentially process multiple data flows. See Process Data Using a Sequence of Data Flows.

To locate the generated dataset, from the Oracle Analytics home page, navigate to Data, then Datasets.

Description of the illustration vision_output_ds.png

For more detail about the generated results, see Output Data Generated for Face Detection, Object Detection, Image Classification, and Text Detection Analysis Models.