In the first instance of setting up Amazon Rekognition will create
You can also create a project on the Projects page. You can access the Projects page via the left navigation pane.
To create your analogue gauge detection model, you first need to create a dataset to train the model with. For this workshop, our dataset is composed of images you have taken of the analogue gauge.
To create your dataset:
You can now review your images to verify your dataset. Add any missing labels, edit labels for images that are labeled incorrectly, and add images and edit labels.
You can label the images by applying bounding boxes on all images with analogue gauges.
In this workshop we are adding custom labels for analogue gauges. Read the AWS blog for training other objects with Amazon Rekognition Custom Labels.
Apply the label to the pressure gauges in the images by selecting all the images with a pressure gauge and choosing Draw bounding box.
You can use the Shift key to automatically select multiple images between the first and last selected images.
After you label your images, you are ready to train your model.
As part of model training, Amazon Rekognition Custom Labels requires a labeled test dataset. Amazon Rekognition Custom Labels uses the test dataset to verify how well your trained model predicts the correct labels and generate evaluation metrics. Images in the test dataset are not used to train your model and should represent the same types of images you will use your model to analyze.
Choose Train new model.
For Choose project, choose your GaugeDetection project.
For Choose training dataset, choose your GaugeImages dataset.
For Create test set, choose Split training dataset. Amazon Rekognition will hold back 20% of the images for testing and use the remaining 80% of the images to train the model.
Our model took approximately 90 minutes to train. The training time required for your model depends on many factors, including the number of images provided in the dataset and the complexity of the model.
When training is complete, Amazon Rekognition Custom Labels outputs key quality metrics including F1 score, precision, recall, and the assumed threshold for each label. For more information about metrics, see Metrics for Evaluating Your Model.
You can also choose View Test Results to see how our model performed on each test image. The following screenshot shows an example of a correctly identified image of pressure gauge during the model testing (true positive).
Your custom gauge detection model is now ready for use. Amazon Rekognition Custom Labels provides the API calls for starting, using and stopping your model; you don’t need to manage any infrastructure.
In the next steps we will show testing the model.